Live AI model pricing & benchmarks

AI Stats

290 LLM models tracked with pricing & performance benchmarks.
Data sourced from ArtificialAnalysis.com & Epoch.ai

Built by Jonathan R Reed to compare model pricing, context limits and performance in one place.

Data Partners

Epoch AI

Top MMLU model

Gemini 3 Pro Preview (high)

89.8%

Cheapest blended price (3:1)

Gemma 3n E4B Instruct

$0.025

Top ECI Score

Gemini 3 Pro Preview

154.4 ECI

Epoch.ai Models

203

84 benchmark runs

Total models

290

From ArtificialAnalysis.com

Avg input price

$1.841

per 1M input tokens

Avg output price

$7.799

per 1M output tokens

Leaderboard

Switch metrics to see different top-5 rankings.

Metric

1

Gemini 3 Pro Preview (high)

Google

89.8%
2

Claude Opus 4.5 (Reasoning)

Anthropic

89.5%
3

Gemini 3 Pro Preview (low)

Google

89.5%
4

Gemini 3 Flash Preview (Reasoning)

Google

89.0%
5

Claude Opus 4.5 (Non-reasoning)

Anthropic

88.9%

All models

Search and scan every tracked model.

Showing 20 of 290 models

Search models

Sort models

GPT-5.2 (xhigh)

OpenAI

Input: $1.750

Output: $14.000

Pricing

Input (1M): $1.750
Output (1M): $14.000
Blended (3:1): $4.813

Performance

Tokens / s: 101.7
TTFT (token): 36.40s
TTFT (answer): 36.40s

Benchmarks

MMLU Pro 87.4%

GPQA 90.3%

HLE 35.4%

AIME —

LiveCodeBench 88.9%

SciCode 52.1%

Math 500 —

AA Indexes

Intelligence

51.1%

Coding

48.7%

Math

99.0%

Claude Opus 4.5 (Reasoning)

Anthropic

Input: $5.000

Output: $25.000

Pricing

Input (1M): $5.000
Output (1M): $25.000
Blended (3:1): $10.000

Performance

Tokens / s: 83.5
TTFT (token): 1.66s
TTFT (answer): 25.60s

Benchmarks

MMLU Pro 89.5%

GPQA 86.6%

HLE 28.4%

AIME —

LiveCodeBench 87.1%

SciCode 49.5%

Math 500 —

AA Indexes

Intelligence

49.6%

Coding

47.8%

Math

91.3%

Gemini 3 Pro Preview (high)

Google

Input: $2.000

Output: $12.000

Pricing

Input (1M): $2.000
Output (1M): $12.000
Blended (3:1): $4.500

Performance

Tokens / s: 116.3
TTFT (token): 31.67s
TTFT (answer): 31.67s

Benchmarks

MMLU Pro 89.8%

GPQA 90.8%

HLE 37.2%

AIME —

LiveCodeBench 91.7%

SciCode 56.1%

Math 500 —

AA Indexes

Intelligence

48.4%

Coding

46.5%

Math

95.7%

GPT-5.1 (high)

OpenAI

Input: $1.250

Output: $10.000

Pricing

Input (1M): $1.250
Output (1M): $10.000
Blended (3:1): $3.438

Performance

Tokens / s: 93.4
TTFT (token): 42.56s
TTFT (answer): 42.56s

Benchmarks

MMLU Pro 87.0%

GPQA 87.3%

HLE 26.5%

AIME —

LiveCodeBench 86.8%

SciCode 43.3%

Math 500 —

AA Indexes

Intelligence

47.5%

Coding

44.7%

Math

94.0%

Gemini 3 Flash Preview (Reasoning)

Google

Input: $0.500

Output: $3.000

Pricing

Input (1M): $0.500
Output (1M): $3.000
Blended (3:1): $1.125

Performance

Tokens / s: 200.6
TTFT (token): 12.96s
TTFT (answer): 12.96s

Benchmarks

MMLU Pro 89.0%

GPQA 89.8%

HLE 34.7%

AIME —

LiveCodeBench 90.8%

SciCode 50.6%

Math 500 —

AA Indexes

Intelligence

46.2%

Coding

42.6%

Math

97.0%

GPT-5.2 (medium)

OpenAI

Input: $1.750

Output: $14.000

Pricing

Input (1M): $1.750
Output (1M): $14.000
Blended (3:1): $4.813

Performance

Tokens / s: —
TTFT (token): —
TTFT (answer): —

Benchmarks

MMLU Pro 85.9%

GPQA 86.4%

HLE 24.9%

AIME —

LiveCodeBench 89.4%

SciCode 46.2%

Math 500 —

AA Indexes

Intelligence

45.8%

Coding

44.2%

Math

96.7%

GPT-5 (high)

OpenAI

Input: $1.250

Output: $10.000

Pricing

Input (1M): $1.250
Output (1M): $10.000
Blended (3:1): $3.438

Performance

Tokens / s: 112.4
TTFT (token): 102.36s
TTFT (answer): 102.36s

Benchmarks

MMLU Pro 87.1%

GPQA 85.4%

HLE 26.5%

AIME 95.7%

LiveCodeBench 84.6%

SciCode 42.9%

Math 500 99.4%

AA Indexes

Intelligence

44.5%

Coding

36.0%

Math

94.3%

GPT-5 Codex (high)

OpenAI

Input: $1.250

Output: $10.000

Pricing

Input (1M): $1.250
Output (1M): $10.000
Blended (3:1): $3.438

Performance

Tokens / s: 156.2
TTFT (token): 22.61s
TTFT (answer): 22.61s

Benchmarks

MMLU Pro 86.5%

GPQA 83.7%

HLE 25.6%

AIME —

LiveCodeBench 84.0%

SciCode 40.9%

Math 500 —

AA Indexes

Intelligence

44.4%

Coding

38.9%

Math

98.7%

Claude Opus 4.5 (Non-reasoning)

Anthropic

Input: $5.000

Output: $25.000

Pricing

Input (1M): $5.000
Output (1M): $25.000
Blended (3:1): $10.000

Performance

Tokens / s: 79.0
TTFT (token): 1.74s
TTFT (answer): 1.74s

Benchmarks

MMLU Pro 88.9%

GPQA 81.0%

HLE 12.9%

AIME —

LiveCodeBench 73.8%

SciCode 47.0%

Math 500 —

AA Indexes

Intelligence

43.0%

Coding

42.9%

Math

62.7%

Claude 4.5 Sonnet (Reasoning)

Anthropic

Input: $3.000

Output: $15.000

Pricing

Input (1M): $3.000
Output (1M): $15.000
Blended (3:1): $6.000

Performance

Tokens / s: 77.6
TTFT (token): 1.48s
TTFT (answer): 27.24s

Benchmarks

MMLU Pro 87.5%

GPQA 83.4%

HLE 17.3%

AIME —

LiveCodeBench 71.4%

SciCode 44.7%

Math 500 —

AA Indexes

Intelligence

42.8%

Coding

38.6%

Math

88.0%

GPT-5 (medium)

OpenAI

Input: $1.250

Output: $10.000

Pricing

Input (1M): $1.250
Output (1M): $10.000
Blended (3:1): $3.438

Performance

Tokens / s: 103.1
TTFT (token): 47.16s
TTFT (answer): 47.16s

Benchmarks

MMLU Pro 86.7%

GPQA 84.2%

HLE 23.5%

AIME 91.7%

LiveCodeBench 70.3%

SciCode 41.1%

Math 500 99.1%

AA Indexes

Intelligence

42.2%

Coding

39.0%

Math

91.7%

GPT-5.1 Codex (high)

OpenAI

Input: $1.250

Output: $10.000

Pricing

Input (1M): $1.250
Output (1M): $10.000
Blended (3:1): $3.438

Performance

Tokens / s: 202.5
TTFT (token): 11.86s
TTFT (answer): 11.86s

Benchmarks

MMLU Pro 86.0%

GPQA 86.0%

HLE 23.4%

AIME —

LiveCodeBench 84.9%

SciCode 40.2%

Math 500 —

AA Indexes

Intelligence

42.0%

Coding

36.6%

Math

95.7%

GLM-4.7 (Reasoning)

Z AI

Input: $0.550

Output: $2.150

Pricing

Input (1M): $0.550
Output (1M): $2.150
Blended (3:1): $0.938

Performance

Tokens / s: 123.4
TTFT (token): 0.68s
TTFT (answer): 16.89s

Benchmarks

MMLU Pro 85.6%

GPQA 85.9%

HLE 25.1%

AIME —

LiveCodeBench 89.4%

SciCode 45.1%

Math 500 —

AA Indexes

Intelligence

42.0%

Coding

36.3%

Math

95.0%

DeepSeek V3.2 (Reasoning)

DeepSeek

Input: $0.280

Output: $0.420

Pricing

Input (1M): $0.280
Output (1M): $0.420
Blended (3:1): $0.315

Performance

Tokens / s: 29.3
TTFT (token): 1.17s
TTFT (answer): 69.36s

Benchmarks

MMLU Pro 86.2%

GPQA 84.0%

HLE 22.2%

AIME —

LiveCodeBench 86.2%

SciCode 38.9%

Math 500 —

AA Indexes

Intelligence

41.6%

Coding

36.7%

Math

92.0%

Grok 4

xAI

Input: $3.000

Output: $15.000

Pricing

Input (1M): $3.000
Output (1M): $15.000
Blended (3:1): $6.000

Performance

Tokens / s: 39.7
TTFT (token): 7.94s
TTFT (answer): 7.94s

Benchmarks

MMLU Pro 86.6%

GPQA 87.7%

HLE 23.9%

AIME 94.3%

LiveCodeBench 81.9%

SciCode 45.7%

Math 500 99.0%

AA Indexes

Intelligence

41.4%

Coding

40.5%

Math

92.7%

GPT-5 mini (high)

OpenAI

Input: $0.250

Output: $2.000

Pricing

Input (1M): $0.250
Output (1M): $2.000
Blended (3:1): $0.688

Performance

Tokens / s: 75.5
TTFT (token): 109.46s
TTFT (answer): 109.46s

Benchmarks

MMLU Pro 83.7%

GPQA 82.8%

HLE 19.7%

AIME —

LiveCodeBench 83.8%

SciCode 39.2%

Math 500 —

AA Indexes

Intelligence

41.0%

Coding

35.3%

Math

90.7%

Gemini 3 Pro Preview (low)

Google

Input: $2.000

Output: $12.000

Pricing

Input (1M): $2.000
Output (1M): $12.000
Blended (3:1): $4.500

Performance

Tokens / s: 126.5
TTFT (token): 4.12s
TTFT (answer): 4.12s

Benchmarks

MMLU Pro 89.5%

GPQA 88.7%

HLE 27.6%

AIME —

LiveCodeBench 85.7%

SciCode 49.9%

Math 500 —

AA Indexes

Intelligence

40.9%

Coding

39.4%

Math

86.7%

o3

OpenAI

Input: $2.000

Output: $8.000

Pricing

Input (1M): $2.000
Output (1M): $8.000
Blended (3:1): $3.500

Performance

Tokens / s: 217.4
TTFT (token): 12.90s
TTFT (answer): 12.90s

Benchmarks

MMLU Pro 85.3%

GPQA 82.7%

HLE 20.0%

AIME 90.3%

LiveCodeBench 80.8%

SciCode 41.0%

Math 500 99.2%

AA Indexes

Intelligence

40.9%

Coding

38.4%

Math

88.3%

o3-pro

OpenAI

Input: $20.000

Output: $80.000

Pricing

Input (1M): $20.000
Output (1M): $80.000
Blended (3:1): $35.000

Performance

Tokens / s: 29.4
TTFT (token): 92.48s
TTFT (answer): 92.48s

Benchmarks

MMLU Pro —

GPQA 84.5%

HLE —

AIME —

LiveCodeBench —

SciCode —

Math 500 —

AA Indexes

Intelligence

40.7%

Coding

—

Math

—

Kimi K2 Thinking

Kimi

Input: $0.600

Output: $2.500

Pricing

Input (1M): $0.600
Output (1M): $2.500
Blended (3:1): $1.075

Performance

Tokens / s: 81.2
TTFT (token): 0.66s
TTFT (answer): 25.29s

Benchmarks

MMLU Pro 84.8%

GPQA 83.8%

HLE 22.3%

AIME —

LiveCodeBench 85.3%

SciCode 42.4%

Math 500 —

AA Indexes

Intelligence

40.6%

Coding

34.8%

Math

94.7%

MiniMax-M2.1

MiniMax

Input: $0.300

Output: $1.200

Pricing

Input (1M): $0.300
Output (1M): $1.200
Blended (3:1): $0.525

Performance

Tokens / s: 70.4
TTFT (token): 1.89s
TTFT (answer): 30.29s

Benchmarks

MMLU Pro 87.5%

GPQA 83.0%

HLE 22.2%

AIME —

LiveCodeBench 81.0%

SciCode 40.7%

Math 500 —

AA Indexes

Intelligence

39.7%

Coding

32.8%

Math

82.7%

MiMo-V2-Flash (Reasoning)

Xiaomi

Input: $0.100

Output: $0.300

Pricing

Input (1M): $0.100
Output (1M): $0.300
Blended (3:1): $0.150

Performance

Tokens / s: 102.1
TTFT (token): 2.76s
TTFT (answer): 22.35s

Benchmarks

MMLU Pro 84.3%

GPQA 84.6%

HLE 21.1%

AIME —

LiveCodeBench 86.8%

SciCode 39.4%

Math 500 —

AA Indexes

Intelligence

39.4%

Coding

31.8%

Math

96.3%

GPT-5 mini (medium)

OpenAI

Input: $0.250

Output: $2.000

Pricing

Input (1M): $0.250
Output (1M): $2.000
Blended (3:1): $0.688

Performance

Tokens / s: 74.8
TTFT (token): 30.71s
TTFT (answer): 30.71s

Benchmarks

MMLU Pro 82.8%

GPQA 80.3%

HLE 14.6%

AIME —

LiveCodeBench 69.2%

SciCode 41.0%

Math 500 —

AA Indexes

Intelligence

38.9%

Coding

32.9%

Math

85.0%

GPT-5 (low)

OpenAI

Input: $1.250

Output: $10.000

Pricing

Input (1M): $1.250
Output (1M): $10.000
Blended (3:1): $3.438

Performance

Tokens / s: 98.9
TTFT (token): 27.90s
TTFT (answer): 27.90s

Benchmarks

MMLU Pro 86.0%

GPQA 80.8%

HLE 18.4%

AIME 83.0%

LiveCodeBench 76.3%

SciCode 39.1%

Math 500 98.7%

AA Indexes

Intelligence

38.9%

Coding

30.7%

Math

83.0%

Claude 4 Sonnet (Reasoning)

Anthropic

Input: $3.000

Output: $15.000

Pricing

Input (1M): $3.000
Output (1M): $15.000
Blended (3:1): $6.000

Performance

Tokens / s: 78.9
TTFT (token): 1.55s
TTFT (answer): 26.90s

Benchmarks

MMLU Pro 84.2%

GPQA 77.7%

HLE 9.6%

AIME 77.3%

LiveCodeBench 65.5%

SciCode 40.0%

Math 500 99.1%

AA Indexes

Intelligence

38.5%

Coding

34.1%

Math

74.3%

GPT-5.1 Codex mini (high)

OpenAI

Input: $0.250

Output: $2.000

Pricing

Input (1M): $0.250
Output (1M): $2.000
Blended (3:1): $0.688

Performance

Tokens / s: 116.5
TTFT (token): 12.27s
TTFT (answer): 12.27s

Benchmarks

MMLU Pro 82.0%

GPQA 81.3%

HLE 16.9%

AIME —

LiveCodeBench 83.6%

SciCode 42.6%

Math 500 —

AA Indexes

Intelligence

38.4%

Coding

36.4%

Math

91.7%

Grok 4.1 Fast (Reasoning)

xAI

Input: $0.200

Output: $0.500

Pricing

Input (1M): $0.200
Output (1M): $0.500
Blended (3:1): $0.275

Performance

Tokens / s: 121.1
TTFT (token): 14.68s
TTFT (answer): 14.68s

Benchmarks

MMLU Pro 85.4%

GPQA 85.3%

HLE 17.6%

AIME —

LiveCodeBench 82.2%

SciCode 44.2%

Math 500 —

AA Indexes

Intelligence

38.4%

Coding

30.9%

Math

89.3%

Claude 4.5 Sonnet (Non-reasoning)

Anthropic

Input: $3.000

Output: $15.000

Pricing

Input (1M): $3.000
Output (1M): $15.000
Blended (3:1): $6.000

Performance

Tokens / s: 54.1
TTFT (token): 1.14s
TTFT (answer): 1.14s

Benchmarks

MMLU Pro 86.0%

GPQA 72.7%

HLE 7.1%

AIME —

LiveCodeBench 59.0%

SciCode 42.8%

Math 500 —

AA Indexes

Intelligence

36.9%

Coding

33.5%

Math

37.0%

Claude 4.5 Haiku (Reasoning)

Anthropic

Input: $1.000

Output: $5.000

Pricing

Input (1M): $1.000
Output (1M): $5.000
Blended (3:1): $2.000

Performance

Tokens / s: 101.2
TTFT (token): 0.72s
TTFT (answer): 20.48s

Benchmarks

MMLU Pro 76.0%

GPQA 67.2%

HLE 9.7%

AIME —

LiveCodeBench 61.5%

SciCode 43.3%

Math 500 —

AA Indexes

Intelligence

36.9%

Coding

32.6%

Math

83.7%

KAT-Coder-Pro V1

KwaiKAT

Input: $0.300

Output: $1.200

Pricing

Input (1M): $0.300
Output (1M): $1.200
Blended (3:1): $0.525

Performance

Tokens / s: 67.0
TTFT (token): 2.07s
TTFT (answer): 2.07s

Benchmarks

MMLU Pro 81.3%

GPQA 76.4%

HLE 33.4%

AIME —

LiveCodeBench 74.7%

SciCode 36.6%

Math 500 —

AA Indexes

Intelligence

36.2%

Coding

18.3%

Math

94.7%

MiniMax-M2

MiniMax

Input: $0.300

Output: $1.200

Pricing

Input (1M): $0.300
Output (1M): $1.200
Blended (3:1): $0.525

Performance

Tokens / s: 85.6
TTFT (token): 1.37s
TTFT (answer): 24.74s

Benchmarks

MMLU Pro 82.0%

GPQA 77.7%

HLE 12.5%

AIME —

LiveCodeBench 82.6%

SciCode 36.1%

Math 500 —

AA Indexes

Intelligence

36.0%

Coding

29.2%

Math

78.3%

Nova 2.0 Pro Preview (medium)

Amazon

Input: $1.250

Output: $10.000

Pricing

Input (1M): $1.250
Output (1M): $10.000
Blended (3:1): $3.438

Performance

Tokens / s: 133.7
TTFT (token): 17.30s
TTFT (answer): 32.27s

Benchmarks

MMLU Pro 83.0%

GPQA 78.5%

HLE 8.9%

AIME —

LiveCodeBench 73.0%

SciCode 42.7%

Math 500 —

AA Indexes

Intelligence

35.6%

Coding

30.4%

Math

89.0%

Gemini 3 Flash Preview (Non-reasoning)

Google

Input: $0.500

Output: $3.000

Pricing

Input (1M): $0.500
Output (1M): $3.000
Blended (3:1): $1.125

Performance

Tokens / s: 167.8
TTFT (token): 0.82s
TTFT (answer): 0.82s

Benchmarks

MMLU Pro 88.2%

GPQA 81.2%

HLE 14.1%

AIME —

LiveCodeBench 79.7%

SciCode 49.9%

Math 500 —

AA Indexes

Intelligence

35.1%

Coding

37.8%

Math

55.7%

Grok 4 Fast (Reasoning)

xAI

Input: $0.200

Output: $0.500

Pricing

Input (1M): $0.200
Output (1M): $0.500
Blended (3:1): $0.275

Performance

Tokens / s: 137.7
TTFT (token): 4.56s
TTFT (answer): 4.56s

Benchmarks

MMLU Pro 85.0%

GPQA 84.7%

HLE 17.0%

AIME —

LiveCodeBench 83.2%

SciCode 44.2%

Math 500 —

AA Indexes

Intelligence

34.8%

Coding

27.4%

Math

89.7%

Claude 3.7 Sonnet (Reasoning)

Anthropic

Input: $3.000

Output: $15.000

Pricing

Input (1M): $3.000
Output (1M): $15.000
Blended (3:1): $6.000

Performance

Tokens / s: —
TTFT (token): —
TTFT (answer): —

Benchmarks

MMLU Pro 83.7%

GPQA 77.2%

HLE 10.3%

AIME 48.7%

LiveCodeBench 47.3%

SciCode 40.3%

Math 500 94.7%

AA Indexes

Intelligence

34.6%

Coding

27.6%

Math

56.3%

Gemini 2.5 Pro

Google

Input: $1.250

Output: $10.000

Pricing

Input (1M): $1.250
Output (1M): $10.000
Blended (3:1): $3.438

Performance

Tokens / s: 152.1
TTFT (token): 36.55s
TTFT (answer): 36.55s

Benchmarks

MMLU Pro 86.2%

GPQA 84.4%

HLE 21.1%

AIME 88.7%

LiveCodeBench 80.1%

SciCode 42.8%

Math 500 96.7%

AA Indexes

Intelligence

34.4%

Coding

31.9%

Math

87.7%

GLM-4.7 (Non-reasoning)

Z AI

Input: $0.550

Output: $2.150

Pricing

Input (1M): $0.550
Output (1M): $2.150
Blended (3:1): $0.938

Performance

Tokens / s: 116.0
TTFT (token): 0.55s
TTFT (answer): 0.55s

Benchmarks

MMLU Pro 79.4%

GPQA 66.4%

HLE 6.1%

AIME —

LiveCodeBench 56.2%

SciCode 35.4%

Math 500 —

AA Indexes

Intelligence

34.1%

Coding

32.0%

Math

48.0%

DeepSeek V3.1 Terminus (Reasoning)

DeepSeek

Input: $0.400

Output: $2.000

Pricing

Input (1M): $0.400
Output (1M): $2.000
Blended (3:1): $0.800

Performance

Tokens / s: —
TTFT (token): —
TTFT (answer): —

Benchmarks

MMLU Pro 85.1%

GPQA 79.2%

HLE 15.2%

AIME —

LiveCodeBench 79.8%

SciCode 40.6%

Math 500 —

AA Indexes

Intelligence

33.8%

Coding

33.7%

Math

89.7%

Doubao Seed Code

ByteDance Seed

Input: $0.170

Output: $1.120

Pricing

Input (1M): $0.170
Output (1M): $1.120
Blended (3:1): $0.407

Performance

Tokens / s: 37.0
TTFT (token): 2.40s
TTFT (answer): 56.40s

Benchmarks

MMLU Pro 85.4%

GPQA 76.4%

HLE 13.3%

AIME —

LiveCodeBench 76.6%

SciCode 40.7%

Math 500 —

AA Indexes

Intelligence

33.5%

Coding

31.3%

Math

79.3%

GPT-5.2 (Non-reasoning)

OpenAI

Input: $1.750

Output: $14.000

Pricing

Input (1M): $1.750
Output (1M): $14.000
Blended (3:1): $4.813

Performance

Tokens / s: 70.1
TTFT (token): 0.50s
TTFT (answer): 0.50s

Benchmarks

MMLU Pro 81.4%

GPQA 71.2%

HLE 7.3%

AIME —

LiveCodeBench 66.9%

SciCode 40.4%

Math 500 —

AA Indexes

Intelligence

33.4%

Coding

34.7%

Math

51.0%

gpt-oss-120B (high)

OpenAI

Input: $0.150

Output: $0.600

Pricing

Input (1M): $0.150
Output (1M): $0.600
Blended (3:1): $0.263

Performance

Tokens / s: 327.4
TTFT (token): 0.49s
TTFT (answer): 6.60s

Benchmarks

MMLU Pro 80.8%

GPQA 78.2%

HLE 18.5%

AIME —

LiveCodeBench 87.8%

SciCode 38.9%

Math 500 —

AA Indexes

Intelligence

33.2%

Coding

28.6%

Math

93.4%

o4-mini (high)

OpenAI

Input: $1.100

Output: $4.400

Pricing

Input (1M): $1.100
Output (1M): $4.400
Blended (3:1): $1.925

Performance

Tokens / s: 132.1
TTFT (token): 50.46s
TTFT (answer): 50.46s

Benchmarks

MMLU Pro 83.2%

GPQA 78.4%

HLE 17.5%

AIME 94.0%

LiveCodeBench 85.9%

SciCode 46.5%

Math 500 98.9%

AA Indexes

Intelligence

33.1%

Coding

25.6%

Math

90.7%

Claude 4 Sonnet (Non-reasoning)

Anthropic

Input: $3.000

Output: $15.000

Pricing

Input (1M): $3.000
Output (1M): $15.000
Blended (3:1): $6.000

Performance

Tokens / s: 72.5
TTFT (token): 1.77s
TTFT (answer): 1.77s

Benchmarks

MMLU Pro 83.7%

GPQA 68.3%

HLE 4.0%

AIME 40.7%

LiveCodeBench 44.9%

SciCode 37.3%

Math 500 93.4%

AA Indexes

Intelligence

33.0%

Coding

30.6%

Math

38.0%

DeepSeek V3.2 Exp (Reasoning)

DeepSeek

Input: $0.280

Output: $0.420

Pricing

Input (1M): $0.280
Output (1M): $0.420
Blended (3:1): $0.315

Performance

Tokens / s: 29.1
TTFT (token): 1.24s
TTFT (answer): 69.95s

Benchmarks

MMLU Pro 85.0%

GPQA 79.7%

HLE 13.8%

AIME —

LiveCodeBench 78.9%

SciCode 37.7%

Math 500 —

AA Indexes

Intelligence

32.8%

Coding

33.3%

Math

87.7%

Grok 3 mini Reasoning (high)

xAI

Input: $0.300

Output: $0.500

Pricing

Input (1M): $0.300
Output (1M): $0.500
Blended (3:1): $0.350

Performance

Tokens / s: 193.2
TTFT (token): 0.66s
TTFT (answer): 11.01s

Benchmarks

MMLU Pro 82.8%

GPQA 79.1%

HLE 11.1%

AIME 93.3%

LiveCodeBench 69.6%

SciCode 40.6%

Math 500 99.2%

AA Indexes

Intelligence

32.6%

Coding

25.2%

Math

84.7%

GLM-4.6 (Reasoning)

Z AI

Input: $0.575

Output: $2.200

Pricing

Input (1M): $0.575
Output (1M): $2.200
Blended (3:1): $0.981

Performance

Tokens / s: 87.6
TTFT (token): 0.69s
TTFT (answer): 23.53s

Benchmarks

MMLU Pro 82.9%

GPQA 78.0%

HLE 13.3%

AIME —

LiveCodeBench 69.5%

SciCode 38.4%

Math 500 —

AA Indexes

Intelligence

32.5%

Coding

29.5%

Math

86.0%

Qwen3 Max Thinking

Alibaba

Input: $1.200

Output: $6.000

Pricing

Input (1M): $1.200
Output (1M): $6.000
Blended (3:1): $2.400

Performance

Tokens / s: 36.8
TTFT (token): 1.73s
TTFT (answer): 56.02s

Benchmarks

MMLU Pro 82.4%

GPQA 77.6%

HLE 12.0%

AIME —

LiveCodeBench 53.5%

SciCode 38.7%

Math 500 —

AA Indexes

Intelligence

32.5%

Coding

24.5%

Math

82.3%

Nova 2.0 Pro Preview (low)

Amazon

Input: $1.250

Output: $10.000

Pricing

Input (1M): $1.250
Output (1M): $10.000
Blended (3:1): $3.438

Performance

Tokens / s: 131.0
TTFT (token): 13.06s
TTFT (answer): 28.33s

Benchmarks

MMLU Pro 82.2%

GPQA 75.1%

HLE 5.2%

AIME —

LiveCodeBench 63.8%

SciCode 38.7%

Math 500 —

AA Indexes

Intelligence

32.4%

Coding

24.5%

Math

63.3%

DeepSeek V3.2 (Non-reasoning)

DeepSeek

Input: $0.280

Output: $0.420

Pricing

Input (1M): $0.280
Output (1M): $0.420
Blended (3:1): $0.315

Performance

Tokens / s: 28.8
TTFT (token): 1.22s
TTFT (answer): 1.22s

Benchmarks

MMLU Pro 83.7%

GPQA 75.1%

HLE 10.5%

AIME —

LiveCodeBench 59.3%

SciCode 38.7%

Math 500 —

AA Indexes

Intelligence

32.2%

Coding

34.6%

Math

59.0%

Claude 4.1 Opus (Reasoning)

Anthropic

Input: $15.000

Output: $75.000

Pricing

Input (1M): $15.000
Output (1M): $75.000
Blended (3:1): $30.000

Performance

Tokens / s: 48.4
TTFT (token): 1.22s
TTFT (answer): 42.55s

Benchmarks

MMLU Pro 88.0%

GPQA 80.9%

HLE 11.9%

AIME —

LiveCodeBench 65.4%

SciCode 40.9%

Math 500 —

AA Indexes

Intelligence

31.9%

Coding

36.5%

Math

80.3%

Qwen3 Max

Alibaba

Input: $1.200

Output: $6.000

Pricing

Input (1M): $1.200
Output (1M): $6.000
Blended (3:1): $2.400

Performance

Tokens / s: 32.9
TTFT (token): 1.97s
TTFT (answer): 1.97s

Benchmarks

MMLU Pro 84.1%

GPQA 76.4%

HLE 11.1%

AIME —

LiveCodeBench 76.7%

SciCode 38.3%

Math 500 —

AA Indexes

Intelligence

31.2%

Coding

26.4%

Math

80.7%

Gemini 2.5 Flash Preview (Sep '25) (Reasoning)

Google

Input: $0.300

Output: $2.500

Pricing

Input (1M): $0.300
Output (1M): $2.500
Blended (3:1): $0.850

Performance

Tokens / s: 290.0
TTFT (token): 12.33s
TTFT (answer): 12.33s

Benchmarks

MMLU Pro 84.2%

GPQA 79.3%

HLE 12.7%

AIME —

LiveCodeBench 71.3%

SciCode 40.5%

Math 500 —

AA Indexes

Intelligence

31.0%

Coding

24.6%

Math

78.3%

Claude 3.7 Sonnet (Non-reasoning)

Anthropic

Input: $3.000

Output: $15.000

Pricing

Input (1M): $3.000
Output (1M): $15.000
Blended (3:1): $6.000

Performance

Tokens / s: —
TTFT (token): —
TTFT (answer): —

Benchmarks

MMLU Pro 80.3%

GPQA 65.6%

HLE 4.8%

AIME 22.3%

LiveCodeBench 39.4%

SciCode 37.6%

Math 500 85.0%

AA Indexes

Intelligence

30.9%

Coding

26.7%

Math

21.0%

Claude 4.5 Haiku (Non-reasoning)

Anthropic

Input: $1.000

Output: $5.000

Pricing

Input (1M): $1.000
Output (1M): $5.000
Blended (3:1): $2.000

Performance

Tokens / s: 92.4
TTFT (token): 0.61s
TTFT (answer): 0.61s

Benchmarks

MMLU Pro 80.0%

GPQA 64.6%

HLE 4.3%

AIME —

LiveCodeBench 51.1%

SciCode 34.4%

Math 500 —

AA Indexes

Intelligence

30.8%

Coding

29.6%

Math

39.0%

o1

OpenAI

Input: $15.000

Output: $60.000

Pricing

Input (1M): $15.000
Output (1M): $60.000
Blended (3:1): $26.250

Performance

Tokens / s: 169.7
TTFT (token): 18.55s
TTFT (answer): 18.55s

Benchmarks

MMLU Pro 84.1%

GPQA 74.7%

HLE 7.7%

AIME 72.3%

LiveCodeBench 67.9%

SciCode 35.8%

Math 500 97.0%

AA Indexes

Intelligence

30.8%

Coding

20.5%

Math

—

MiMo-V2-Flash (Non-reasoning)

Xiaomi

Input: $0.100

Output: $0.300

Pricing

Input (1M): $0.100
Output (1M): $0.300
Blended (3:1): $0.150

Performance

Tokens / s: 92.0
TTFT (token): 2.01s
TTFT (answer): 2.01s

Benchmarks

MMLU Pro 74.4%

GPQA 65.6%

HLE 8.0%

AIME —

LiveCodeBench 40.2%

SciCode 25.9%

Math 500 —

AA Indexes

Intelligence

30.4%

Coding

25.8%

Math

67.7%

Gemini 2.5 Pro Preview (Mar' 25)

Google

Input: $1.250

Output: $10.000

Pricing

Input (1M): $1.250
Output (1M): $10.000
Blended (3:1): $3.438

Performance

Tokens / s: —
TTFT (token): —
TTFT (answer): —

Benchmarks

MMLU Pro 85.8%

GPQA 83.6%

HLE 17.1%

AIME 87.0%

LiveCodeBench 77.8%

SciCode 39.5%

Math 500 98.0%

AA Indexes

Intelligence

30.3%

Coding

46.7%

Math

—

GLM-4.6 (Non-reasoning)

Z AI

Input: $0.600

Output: $2.200

Pricing

Input (1M): $0.600
Output (1M): $2.200
Blended (3:1): $1.000

Performance

Tokens / s: 87.2
TTFT (token): 1.63s
TTFT (answer): 1.63s

Benchmarks

MMLU Pro 78.4%

GPQA 63.2%

HLE 5.2%

AIME —

LiveCodeBench 56.1%

SciCode 33.1%

Math 500 —

AA Indexes

Intelligence

30.1%

Coding

30.2%

Math

44.3%

Nova 2.0 Lite (medium)

Amazon

Input: $0.300

Output: $2.500

Pricing

Input (1M): $0.300
Output (1M): $2.500
Blended (3:1): $0.850

Performance

Tokens / s: 240.2
TTFT (token): 12.65s
TTFT (answer): 20.97s

Benchmarks

MMLU Pro 81.3%

GPQA 76.8%

HLE 8.6%

AIME —

LiveCodeBench 66.3%

SciCode 36.8%

Math 500 —

AA Indexes

Intelligence

30.1%

Coding

23.9%

Math

88.7%

Gemini 2.5 Pro Preview (May' 25)

Google

Input: $1.250

Output: $10.000

Pricing

Input (1M): $1.250
Output (1M): $10.000
Blended (3:1): $3.438

Performance

Tokens / s: —
TTFT (token): —
TTFT (answer): —

Benchmarks

MMLU Pro 83.7%

GPQA 82.2%

HLE 15.4%

AIME 84.3%

LiveCodeBench 77.0%

SciCode 41.6%

Math 500 98.6%

AA Indexes

Intelligence

29.5%

Coding

—

Math

—

Qwen3 235B A22B 2507 (Reasoning)

Alibaba

Input: $0.700

Output: $8.400

Pricing

Input (1M): $0.700
Output (1M): $8.400
Blended (3:1): $2.625

Performance

Tokens / s: 75.9
TTFT (token): 1.08s
TTFT (answer): 27.43s

Benchmarks

MMLU Pro 84.3%

GPQA 79.0%

HLE 15.0%

AIME 94.0%

LiveCodeBench 78.8%

SciCode 42.4%

Math 500 98.4%

AA Indexes

Intelligence

29.5%

Coding

23.2%

Math

91.0%

DeepSeek V3.2 Speciale

DeepSeek

Input: $0.400

Output: $0.500

Pricing

Input (1M): $0.400
Output (1M): $0.500
Blended (3:1): $0.425

Performance

Tokens / s: —
TTFT (token): —
TTFT (answer): —

Benchmarks

MMLU Pro 86.3%

GPQA 87.1%

HLE 26.1%

AIME —

LiveCodeBench 89.6%

SciCode 44.0%

Math 500 —

AA Indexes

Intelligence

29.4%

Coding

37.9%

Math

96.7%

ERNIE 5.0 Thinking Preview

Baidu

Input: $0.840

Output: $3.370

Pricing

Input (1M): $0.840
Output (1M): $3.370
Blended (3:1): $1.472

Performance

Tokens / s: 25.6
TTFT (token): 4.07s
TTFT (answer): 82.15s

Benchmarks

MMLU Pro 83.0%

GPQA 77.7%

HLE 12.7%

AIME —

LiveCodeBench 81.2%

SciCode 37.5%

Math 500 —

AA Indexes

Intelligence

29.2%

Coding

29.2%

Math

85.0%

Qwen3 VL 32B (Reasoning)

Alibaba

Input: $0.700

Output: $8.400

Pricing

Input (1M): $0.700
Output (1M): $8.400
Blended (3:1): $2.625

Performance

Tokens / s: 50.5
TTFT (token): 0.98s
TTFT (answer): 40.55s

Benchmarks

MMLU Pro 81.8%

GPQA 73.3%

HLE 9.6%

AIME —

LiveCodeBench 73.8%

SciCode 28.5%

Math 500 —

AA Indexes

Intelligence

28.6%

Coding

14.5%

Math

84.7%

DeepSeek V3.1 Terminus (Non-reasoning)

DeepSeek

Input: $0.400

Output: $1.680

Pricing

Input (1M): $0.400
Output (1M): $1.680
Blended (3:1): $0.800

Performance

Tokens / s: —
TTFT (token): —
TTFT (answer): —

Benchmarks

MMLU Pro 83.6%

GPQA 75.1%

HLE 8.4%

AIME —

LiveCodeBench 52.9%

SciCode 32.1%

Math 500 —

AA Indexes

Intelligence

28.3%

Coding

31.9%

Math

53.7%

DeepSeek V3.2 Exp (Non-reasoning)

DeepSeek

Input: $0.280

Output: $0.420

Pricing

Input (1M): $0.280
Output (1M): $0.420
Blended (3:1): $0.315

Performance

Tokens / s: 29.4
TTFT (token): 1.20s
TTFT (answer): 1.20s

Benchmarks

MMLU Pro 83.6%

GPQA 73.8%

HLE 8.6%

AIME —

LiveCodeBench 55.4%

SciCode 39.9%

Math 500 —

AA Indexes

Intelligence

28.3%

Coding

30.0%

Math

57.7%

MiniMax-Text-01

MiniMax

Input: $0.200

Output: $1.100

Pricing

Input (1M): $0.200
Output (1M): $1.100
Blended (3:1): $0.425

Performance

Tokens / s: 28.5
TTFT (token): 1.32s
TTFT (answer): 1.32s

Benchmarks

MMLU Pro 75.9%

GPQA 57.8%

HLE 4.2%

AIME 13.0%

LiveCodeBench 24.7%

SciCode 25.0%

Math 500 75.3%

AA Indexes

Intelligence

28.3%

Coding

17.3%

Math

12.3%

Kimi K2 0905

Kimi

Input: $0.990

Output: $2.500

Pricing

Input (1M): $0.990
Output (1M): $2.500
Blended (3:1): $1.200

Performance

Tokens / s: 47.2
TTFT (token): 0.54s
TTFT (answer): 0.54s

Benchmarks

MMLU Pro 81.9%

GPQA 76.7%

HLE 6.3%

AIME —

LiveCodeBench 61.0%

SciCode 30.7%

Math 500 —

AA Indexes

Intelligence

28.1%

Coding

25.9%

Math

57.3%

DeepSeek V3.1 (Reasoning)

DeepSeek

Input: $0.590

Output: $1.690

Pricing

Input (1M): $0.590
Output (1M): $1.690
Blended (3:1): $0.865

Performance

Tokens / s: —
TTFT (token): —
TTFT (answer): —

Benchmarks

MMLU Pro 85.1%

GPQA 77.9%

HLE 13.0%

AIME —

LiveCodeBench 78.4%

SciCode 39.1%

Math 500 —

AA Indexes

Intelligence

28.1%

Coding

29.7%

Math

89.7%

DeepSeek V3.1 (Non-reasoning)

DeepSeek

Input: $0.560

Output: $1.680

Pricing

Input (1M): $0.560
Output (1M): $1.680
Blended (3:1): $0.840

Performance

Tokens / s: —
TTFT (token): —
TTFT (answer): —

Benchmarks

MMLU Pro 83.3%

GPQA 73.5%

HLE 6.3%

AIME —

LiveCodeBench 57.7%

SciCode 36.7%

Math 500 —

AA Indexes

Intelligence

27.9%

Coding

28.4%

Math

49.7%

Nova 2.0 Omni (medium)

Amazon

Input: $0.300

Output: $2.500

Pricing

Input (1M): $0.300
Output (1M): $2.500
Blended (3:1): $0.850

Performance

Tokens / s: —
TTFT (token): —
TTFT (answer): —

Benchmarks

MMLU Pro 80.9%

GPQA 76.0%

HLE 6.8%

AIME —

LiveCodeBench 66.0%

SciCode 36.2%

Math 500 —

AA Indexes

Intelligence

27.9%

Coding

15.1%

Math

89.7%

Qwen3 VL 235B A22B (Reasoning)

Alibaba

Input: $0.700

Output: $8.400

Pricing

Input (1M): $0.700
Output (1M): $8.400
Blended (3:1): $2.625

Performance

Tokens / s: 43.6
TTFT (token): 1.09s
TTFT (answer): 46.98s

Benchmarks

MMLU Pro 83.6%

GPQA 77.2%

HLE 10.1%

AIME —

LiveCodeBench 64.6%

SciCode 39.9%

Math 500 —

AA Indexes

Intelligence

27.6%

Coding

20.9%

Math

88.3%

Magistral Medium 1.2

Mistral

Input: $2.000

Output: $5.000

Pricing

Input (1M): $2.000
Output (1M): $5.000
Blended (3:1): $2.750

Performance

Tokens / s: 35.8
TTFT (token): 0.52s
TTFT (answer): 56.37s

Benchmarks

MMLU Pro 81.5%

GPQA 73.9%

HLE 9.6%

AIME —

LiveCodeBench 75.0%

SciCode 39.2%

Math 500 —

AA Indexes

Intelligence

27.5%

Coding

21.7%

Math

82.0%

Claude 4 Opus (Reasoning)

Anthropic

Input: $15.000

Output: $75.000

Pricing

Input (1M): $15.000
Output (1M): $75.000
Blended (3:1): $30.000

Performance

Tokens / s: 48.8
TTFT (token): 1.13s
TTFT (answer): 42.13s

Benchmarks

MMLU Pro 87.3%

GPQA 79.6%

HLE 11.7%

AIME 75.7%

LiveCodeBench 63.6%

SciCode 39.8%

Math 500 98.2%

AA Indexes

Intelligence

27.4%

Coding

34.0%

Math

73.3%

Gemini 2.5 Flash (Reasoning)

Google

Input: $0.300

Output: $2.500

Pricing

Input (1M): $0.300
Output (1M): $2.500
Blended (3:1): $0.850

Performance

Tokens / s: 262.3
TTFT (token): 17.46s
TTFT (answer): 17.46s

Benchmarks

MMLU Pro 83.2%

GPQA 79.0%

HLE 11.1%

AIME 82.3%

LiveCodeBench 69.5%

SciCode 39.4%

Math 500 98.1%

AA Indexes

Intelligence

27.4%

Coding

22.2%

Math

73.3%

GPT-5.1 (Non-reasoning)

OpenAI

Input: $1.250

Output: $10.000

Pricing

Input (1M): $1.250
Output (1M): $10.000
Blended (3:1): $3.438

Performance

Tokens / s: 70.4
TTFT (token): 0.76s
TTFT (answer): 0.76s

Benchmarks

MMLU Pro 80.1%

GPQA 64.3%

HLE 5.2%

AIME —

LiveCodeBench 49.4%

SciCode 36.5%

Math 500 —

AA Indexes

Intelligence

27.4%

Coding

27.3%

Math

38.0%

DeepSeek R1 0528 (May '25)

DeepSeek

Input: $1.350

Output: $4.200

Pricing

Input (1M): $1.350
Output (1M): $4.200
Blended (3:1): $2.362

Performance

Tokens / s: —
TTFT (token): —
TTFT (answer): —

Benchmarks

MMLU Pro 84.9%

GPQA 81.3%

HLE 14.9%

AIME 89.3%

LiveCodeBench 77.0%

SciCode 40.3%

Math 500 98.3%

AA Indexes

Intelligence

27.2%

Coding

24.0%

Math

76.0%

GLM-4.5 (Reasoning)

Z AI

Input: $0.600

Output: $2.200

Pricing

Input (1M): $0.600
Output (1M): $2.200
Blended (3:1): $1.000

Performance

Tokens / s: 50.2
TTFT (token): 0.60s
TTFT (answer): 40.48s

Benchmarks

MMLU Pro 83.5%

GPQA 78.2%

HLE 12.2%

AIME 87.3%

LiveCodeBench 73.8%

SciCode 34.8%

Math 500 97.9%

AA Indexes

Intelligence

26.9%

Coding

26.3%

Math

73.7%

GPT-5 nano (high)

OpenAI

Input: $0.050

Output: $0.400

Pricing

Input (1M): $0.050
Output (1M): $0.400
Blended (3:1): $0.138

Performance

Tokens / s: 115.1
TTFT (token): 138.34s
TTFT (answer): 138.34s

Benchmarks

MMLU Pro 78.0%

GPQA 67.6%

HLE 8.2%

AIME —

LiveCodeBench 78.9%

SciCode 36.6%

Math 500 —

AA Indexes

Intelligence

26.9%

Coding

20.3%

Math

83.7%

Qwen3 Next 80B A3B (Reasoning)

Alibaba

Input: $0.500

Output: $6.000

Pricing

Input (1M): $0.500
Output (1M): $6.000
Blended (3:1): $1.875

Performance

Tokens / s: 158.4
TTFT (token): 1.06s
TTFT (answer): 13.69s

Benchmarks

MMLU Pro 82.4%

GPQA 75.9%

HLE 11.7%

AIME —

LiveCodeBench 78.4%

SciCode 38.8%

Math 500 —

AA Indexes

Intelligence

26.6%

Coding

19.5%

Math

84.3%

Grok Code Fast 1

xAI

Input: $0.200

Output: $1.500

Pricing

Input (1M): $0.200
Output (1M): $1.500
Blended (3:1): $0.525

Performance

Tokens / s: 229.5
TTFT (token): 9.06s
TTFT (answer): 9.06s

Benchmarks

MMLU Pro 79.3%

GPQA 72.7%

HLE 7.5%

AIME —

LiveCodeBench 65.7%

SciCode 36.2%

Math 500 —

AA Indexes

Intelligence

26.2%

Coding

23.7%

Math

43.3%

Qwen3 Max (Preview)

Alibaba

Input: $1.200

Output: $6.000

Pricing

Input (1M): $1.200
Output (1M): $6.000
Blended (3:1): $2.400

Performance

Tokens / s: 32.2
TTFT (token): 1.73s
TTFT (answer): 1.73s

Benchmarks

MMLU Pro 83.8%

GPQA 76.4%

HLE 9.3%

AIME —

LiveCodeBench 65.1%

SciCode 37.0%

Math 500 —

AA Indexes

Intelligence

26.1%

Coding

25.5%

Math

75.0%

GPT-5 nano (medium)

OpenAI

Input: $0.050

Output: $0.400

Pricing

Input (1M): $0.050
Output (1M): $0.400
Blended (3:1): $0.138

Performance

Tokens / s: 126.3
TTFT (token): 59.41s
TTFT (answer): 59.41s

Benchmarks

MMLU Pro 77.2%

GPQA 67.0%

HLE 7.6%

AIME —

LiveCodeBench 76.3%

SciCode 33.8%

Math 500 —

AA Indexes

Intelligence

26.0%

Coding

22.9%

Math

78.3%

o3-mini

OpenAI

Input: $1.100

Output: $4.400

Pricing

Input (1M): $1.100
Output (1M): $4.400
Blended (3:1): $1.925

Performance

Tokens / s: 142.7
TTFT (token): 17.29s
TTFT (answer): 17.29s

Benchmarks

MMLU Pro 79.1%

GPQA 74.8%

HLE 8.7%

AIME 77.0%

LiveCodeBench 71.7%

SciCode 39.9%

Math 500 97.3%

AA Indexes

Intelligence

25.9%

Coding

17.9%

Math

—

Kimi K2

Kimi

Input: $0.600

Output: $2.500

Pricing

Input (1M): $0.600
Output (1M): $2.500
Blended (3:1): $1.075

Performance

Tokens / s: 37.3
TTFT (token): 0.68s
TTFT (answer): 0.68s

Benchmarks

MMLU Pro 82.4%

GPQA 76.6%

HLE 7.0%

AIME 69.3%

LiveCodeBench 55.6%

SciCode 34.5%

Math 500 97.1%

AA Indexes

Intelligence

25.9%

Coding

22.1%

Math

57.0%

GPT-4.1

OpenAI

Input: $2.000

Output: $8.000

Pricing

Input (1M): $2.000
Output (1M): $8.000
Blended (3:1): $3.500

Performance

Tokens / s: 76.9
TTFT (token): 0.52s
TTFT (answer): 0.52s

Benchmarks

MMLU Pro 80.6%

GPQA 66.6%

HLE 4.6%

AIME 43.7%

LiveCodeBench 45.7%

SciCode 38.1%

Math 500 91.3%

AA Indexes

Intelligence

25.9%

Coding

21.8%

Math

34.7%

o1-pro

OpenAI

Input: $150.000

Output: $600.000

Pricing

Input (1M): $150.000
Output (1M): $600.000
Blended (3:1): $262.500

Performance

Tokens / s: —
TTFT (token): —
TTFT (answer): —

Benchmarks

MMLU Pro —

GPQA —

HLE —

AIME —

LiveCodeBench —

SciCode —

Math 500 —

AA Indexes

Intelligence

25.8%

Coding

—

Math

—

Gemini 2.5 Flash Preview (Sep '25) (Non-reasoning)

Google

Input: $0.300

Output: $2.500

Pricing

Input (1M): $0.300
Output (1M): $2.500
Blended (3:1): $0.850

Performance

Tokens / s: 249.2
TTFT (token): 0.30s
TTFT (answer): 0.30s

Benchmarks

MMLU Pro 83.6%

GPQA 76.6%

HLE 7.8%

AIME —

LiveCodeBench 62.5%

SciCode 37.5%

Math 500 —

AA Indexes

Intelligence

25.7%

Coding

22.1%

Math

56.7%

Grok 3

xAI

Input: $3.000

Output: $15.000

Pricing

Input (1M): $3.000
Output (1M): $15.000
Blended (3:1): $6.000

Performance

Tokens / s: 41.9
TTFT (token): 0.77s
TTFT (answer): 0.77s

Benchmarks

MMLU Pro 79.9%

GPQA 69.3%

HLE 5.1%

AIME 33.0%

LiveCodeBench 42.5%

SciCode 36.8%

Math 500 87.0%

AA Indexes

Intelligence

25.4%

Coding

19.8%

Math

58.0%

o3-mini (high)

OpenAI

Input: $1.100

Output: $4.400

Pricing

Input (1M): $1.100
Output (1M): $4.400
Blended (3:1): $1.925

Performance

Tokens / s: 159.1
TTFT (token): 56.06s
TTFT (answer): 56.06s

Benchmarks

MMLU Pro 80.2%

GPQA 77.3%

HLE 12.3%

AIME 86.0%

LiveCodeBench 73.4%

SciCode 39.8%

Math 500 98.5%

AA Indexes

Intelligence

25.1%

Coding

17.3%

Math

—

Seed-OSS-36B-Instruct

ByteDance Seed

Input: $0.210

Output: $0.570

Pricing

Input (1M): $0.210
Output (1M): $0.570
Blended (3:1): $0.300

Performance

Tokens / s: 29.4
TTFT (token): 1.73s
TTFT (answer): 69.67s

Benchmarks

MMLU Pro 81.5%

GPQA 72.6%

HLE 9.1%

AIME —

LiveCodeBench 76.5%

SciCode 36.5%

Math 500 —

AA Indexes

Intelligence

25.1%

Coding

16.7%

Math

84.7%

Nova 2.0 Lite (low)

Amazon

Input: $0.300

Output: $2.500

Pricing

Input (1M): $0.300
Output (1M): $2.500
Blended (3:1): $0.850

Performance

Tokens / s: 250.8
TTFT (token): 6.82s
TTFT (answer): 14.80s

Benchmarks

MMLU Pro 78.8%

GPQA 69.8%

HLE 4.2%

AIME —

LiveCodeBench 46.9%

SciCode 33.3%

Math 500 —

AA Indexes

Intelligence

25.0%

Coding

13.6%

Math

46.7%

Qwen3 Coder 480B A35B Instruct

Alibaba

Input: $1.500

Output: $7.500

Pricing

Input (1M): $1.500
Output (1M): $7.500
Blended (3:1): $3.000

Performance

Tokens / s: 40.5
TTFT (token): 1.69s
TTFT (answer): 1.69s

Benchmarks

MMLU Pro 78.8%

GPQA 61.8%

HLE 4.4%

AIME 47.7%

LiveCodeBench 58.5%

SciCode 35.9%

Math 500 94.2%

AA Indexes

Intelligence

25.0%

Coding

24.6%

Math

39.3%

NVIDIA Nemotron 3 Nano 30B A3B (Reasoning)

NVIDIA

Input: $0.060

Output: $0.240

Pricing

Input (1M): $0.060
Output (1M): $0.240
Blended (3:1): $0.105

Performance

Tokens / s: 191.4
TTFT (token): 0.39s
TTFT (answer): 10.83s

Benchmarks

MMLU Pro 79.4%

GPQA 75.7%

HLE 10.2%

AIME —

LiveCodeBench 74.1%

SciCode 29.6%

Math 500 —

AA Indexes

Intelligence

24.9%

Coding

19.0%

Math

91.0%

gpt-oss-20B (high)

OpenAI

Input: $0.070

Output: $0.200

Pricing

Input (1M): $0.070
Output (1M): $0.200
Blended (3:1): $0.100

Performance

Tokens / s: 307.3
TTFT (token): 0.53s
TTFT (answer): 7.04s

Benchmarks

MMLU Pro 74.8%

GPQA 68.8%

HLE 9.8%

AIME —

LiveCodeBench 77.7%

SciCode 34.4%

Math 500 —

AA Indexes

Intelligence

24.8%

Coding

18.5%

Math

89.3%

Qwen3 235B A22B 2507 Instruct

Alibaba

Input: $0.700

Output: $2.800

Pricing

Input (1M): $0.700
Output (1M): $2.800
Blended (3:1): $1.225

Performance

Tokens / s: 55.4
TTFT (token): 1.04s
TTFT (answer): 1.04s

Benchmarks

MMLU Pro 82.8%

GPQA 75.3%

HLE 10.6%

AIME 71.7%

LiveCodeBench 52.4%

SciCode 36.0%

Math 500 98.0%

AA Indexes

Intelligence

24.7%

Coding

22.1%

Math

71.7%

GPT-5 (minimal)

OpenAI

Input: $1.250

Output: $10.000

Pricing

Input (1M): $1.250
Output (1M): $10.000
Blended (3:1): $3.438

Performance

Tokens / s: 75.0
TTFT (token): 0.82s
TTFT (answer): 0.82s

Benchmarks

MMLU Pro 80.6%

GPQA 67.3%

HLE 5.4%

AIME 36.7%

LiveCodeBench 55.8%

SciCode 38.8%

Math 500 86.1%

AA Indexes

Intelligence

24.6%

Coding

25.1%

Math

31.7%

MiniMax M1 80k

MiniMax

Input: $0.550

Output: $2.200

Pricing

Input (1M): $0.550
Output (1M): $2.200
Blended (3:1): $0.963

Performance

Tokens / s: —
TTFT (token): —
TTFT (answer): —

Benchmarks

MMLU Pro 81.6%

GPQA 69.7%

HLE 8.2%

AIME 84.7%

LiveCodeBench 71.1%

SciCode 37.4%

Math 500 98.0%

AA Indexes

Intelligence

24.4%

Coding

14.5%

Math

61.0%

Nova 2.0 Omni (low)

Amazon

Input: $0.300

Output: $2.500

Pricing

Input (1M): $0.300
Output (1M): $2.500
Blended (3:1): $0.850

Performance

Tokens / s: —
TTFT (token): —
TTFT (answer): —

Benchmarks

MMLU Pro 79.8%

GPQA 69.9%

HLE 4.0%

AIME —

LiveCodeBench 59.2%

SciCode 34.3%

Math 500 —

AA Indexes

Intelligence

24.1%

Coding

13.9%

Math

56.0%

GLM-4.6V (Reasoning)

Z AI

Input: $0.300

Output: $0.900

Pricing

Input (1M): $0.300
Output (1M): $0.900
Blended (3:1): $0.450

Performance

Tokens / s: 68.0
TTFT (token): 0.72s
TTFT (answer): 30.13s

Benchmarks

MMLU Pro 79.9%

GPQA 71.9%

HLE 8.9%

AIME —

LiveCodeBench 16.0%

SciCode 30.4%

Math 500 —

AA Indexes

Intelligence

24.0%

Coding

19.7%

Math

85.3%

GLM-4.5-Air

Z AI

Input: $0.200

Output: $1.100

Pricing

Input (1M): $0.200
Output (1M): $1.100
Blended (3:1): $0.425

Performance

Tokens / s: 98.9
TTFT (token): 0.61s
TTFT (answer): 20.84s

Benchmarks

MMLU Pro 81.5%

GPQA 73.3%

HLE 6.8%

AIME 67.3%

LiveCodeBench 68.4%

SciCode 30.6%

Math 500 96.5%

AA Indexes

Intelligence

23.8%

Coding

23.8%

Math

80.7%

gpt-oss-120B (low)

OpenAI

Input: $0.150

Output: $0.595

Pricing

Input (1M): $0.150
Output (1M): $0.595
Blended (3:1): $0.263

Performance

Tokens / s: 299.1
TTFT (token): 0.48s
TTFT (answer): 7.17s

Benchmarks

MMLU Pro 77.5%

GPQA 67.2%

HLE 5.2%

AIME —

LiveCodeBench 70.7%

SciCode 36.0%

Math 500 —

AA Indexes

Intelligence

23.8%

Coding

15.5%

Math

66.7%

Grok 4.1 Fast (Non-reasoning)

xAI

Input: $0.200

Output: $0.500

Pricing

Input (1M): $0.200
Output (1M): $0.500
Blended (3:1): $0.275

Performance

Tokens / s: 81.1
TTFT (token): 0.69s
TTFT (answer): 0.69s

Benchmarks

MMLU Pro 74.3%

GPQA 63.7%

HLE 5.0%

AIME —

LiveCodeBench 39.9%

SciCode 29.6%

Math 500 —

AA Indexes

Intelligence

23.8%

Coding

19.5%

Math

34.3%

Nova 2.0 Pro Preview (Non-reasoning)

Amazon

Input: $1.250

Output: $10.000

Pricing

Input (1M): $1.250
Output (1M): $10.000
Blended (3:1): $3.438

Performance

Tokens / s: 159.0
TTFT (token): 0.46s
TTFT (answer): 0.46s

Benchmarks

MMLU Pro 77.2%

GPQA 63.6%

HLE 4.0%

AIME —

LiveCodeBench 47.3%

SciCode 28.1%

Math 500 —

AA Indexes

Intelligence

23.7%

Coding

20.5%

Math

30.7%

o1-preview

OpenAI

Input: $16.500

Output: $66.000

Pricing

Input (1M): $16.500
Output (1M): $66.000
Blended (3:1): $28.875

Performance

Tokens / s: —
TTFT (token): —
TTFT (answer): —

Benchmarks

MMLU Pro —

GPQA —

HLE —

AIME —

LiveCodeBench —

SciCode —

Math 500 92.4%

AA Indexes

Intelligence

23.7%

Coding

34.0%

Math

—

Claude 4.1 Opus (Non-reasoning)

Anthropic

Input: $15.000

Output: $75.000

Pricing

Input (1M): $15.000
Output (1M): $75.000
Blended (3:1): $30.000

Performance

Tokens / s: 37.9
TTFT (token): 1.30s
TTFT (answer): 1.30s

Benchmarks

MMLU Pro —

GPQA —

HLE —

AIME —

LiveCodeBench —

SciCode —

Math 500 —

AA Indexes

Intelligence

23.6%

Coding

—

Math

—

GPT-4.1 mini

OpenAI

Input: $0.400

Output: $1.600

Pricing

Input (1M): $0.400
Output (1M): $1.600
Blended (3:1): $0.700

Performance

Tokens / s: 67.2
TTFT (token): 0.45s
TTFT (answer): 0.45s

Benchmarks

MMLU Pro 78.1%

GPQA 66.4%

HLE 4.6%

AIME 43.0%

LiveCodeBench 48.3%

SciCode 40.4%

Math 500 92.5%

AA Indexes

Intelligence

23.0%

Coding

18.5%

Math

46.3%

Qwen3 30B A3B 2507 (Reasoning)

Alibaba

Input: $0.200

Output: $2.400

Pricing

Input (1M): $0.200
Output (1M): $2.400
Blended (3:1): $0.750

Performance

Tokens / s: 165.3
TTFT (token): 1.02s
TTFT (answer): 13.11s

Benchmarks

MMLU Pro 80.5%

GPQA 70.7%

HLE 9.8%

AIME 90.7%

LiveCodeBench 70.7%

SciCode 33.3%

Math 500 97.6%

AA Indexes

Intelligence

22.9%

Coding

14.7%

Math

56.3%

Grok 4 Fast (Non-reasoning)

xAI

Input: $0.200

Output: $0.500

Pricing

Input (1M): $0.200
Output (1M): $0.500
Blended (3:1): $0.275

Performance

Tokens / s: 127.6
TTFT (token): 0.54s
TTFT (answer): 0.54s

Benchmarks

MMLU Pro 73.0%

GPQA 60.6%

HLE 5.0%

AIME —

LiveCodeBench 40.1%

SciCode 32.9%

Math 500 —

AA Indexes

Intelligence

22.9%

Coding

19.0%

Math

41.3%

DeepSeek V3 0324

DeepSeek

Input: $1.195

Output: $1.350

Pricing

Input (1M): $1.195
Output (1M): $1.350
Blended (3:1): $1.250

Performance

Tokens / s: —
TTFT (token): —
TTFT (answer): —

Benchmarks

MMLU Pro 81.9%

GPQA 65.5%

HLE 5.2%

AIME 52.0%

LiveCodeBench 40.5%

SciCode 35.8%

Math 500 94.2%

AA Indexes

Intelligence

22.8%

Coding

22.0%

Math

41.0%

Ring-1T

InclusionAI

Input: $0.560

Output: $2.240

Pricing

Input (1M): $0.560
Output (1M): $2.240
Blended (3:1): $0.980

Performance

Tokens / s: 57.9
TTFT (token): 1.74s
TTFT (answer): 36.31s

Benchmarks

MMLU Pro 80.6%

GPQA 77.4%

HLE 10.2%

AIME —

LiveCodeBench 64.3%

SciCode 36.7%

Math 500 —

AA Indexes

Intelligence

22.7%

Coding

16.8%

Math

89.3%

Mistral Large 3

Mistral

Input: $0.500

Output: $1.500

Pricing

Input (1M): $0.500
Output (1M): $1.500
Blended (3:1): $0.750

Performance

Tokens / s: 50.9
TTFT (token): 0.55s
TTFT (answer): 0.55s

Benchmarks

MMLU Pro 80.7%

GPQA 68.0%

HLE 4.1%

AIME —

LiveCodeBench 46.5%

SciCode 36.2%

Math 500 —

AA Indexes

Intelligence

22.7%

Coding

22.7%

Math

38.0%

Magistral Small 1.2

Mistral

Input: $0.500

Output: $1.500

Pricing

Input (1M): $0.500
Output (1M): $1.500
Blended (3:1): $0.750

Performance

Tokens / s: 192.7
TTFT (token): 0.35s
TTFT (answer): 10.73s

Benchmarks

MMLU Pro 76.8%

GPQA 66.3%

HLE 6.1%

AIME —

LiveCodeBench 72.3%

SciCode 35.2%

Math 500 —

AA Indexes

Intelligence

22.5%

Coding

14.8%

Math

80.3%

Gemini 2.5 Flash-Lite Preview (Sep '25) (Reasoning)

Google

Input: $0.100

Output: $0.400

Pricing

Input (1M): $0.100
Output (1M): $0.400
Blended (3:1): $0.175

Performance

Tokens / s: 576.3
TTFT (token): 6.52s
TTFT (answer): 6.52s

Benchmarks

MMLU Pro 80.8%

GPQA 70.9%

HLE 6.6%

AIME —

LiveCodeBench 68.8%

SciCode 28.7%

Math 500 —

AA Indexes

Intelligence

22.5%

Coding

18.1%

Math

68.7%

INTELLECT-3

Prime Intellect

Input: $0.200

Output: $1.100

Pricing

Input (1M): $0.200
Output (1M): $1.100
Blended (3:1): $0.425

Performance

Tokens / s: 84.8
TTFT (token): 0.59s
TTFT (answer): 24.18s

Benchmarks

MMLU Pro 82.2%

GPQA 76.1%

HLE 12.1%

AIME —

LiveCodeBench 77.7%

SciCode 39.1%

Math 500 —

AA Indexes

Intelligence

22.3%

Coding

19.1%

Math

88.0%

Claude 4 Opus (Non-reasoning)

Anthropic

Input: $15.000

Output: $75.000

Pricing

Input (1M): $15.000
Output (1M): $75.000
Blended (3:1): $30.000

Performance

Tokens / s: 40.2
TTFT (token): 1.18s
TTFT (answer): 1.18s

Benchmarks

MMLU Pro 86.0%

GPQA 70.1%

HLE 5.9%

AIME 56.3%

LiveCodeBench 54.2%

SciCode 40.9%

Math 500 94.1%

AA Indexes

Intelligence

22.2%

Coding

—

Math

36.3%

CompactifAI Llama 4 Scout Slim

Multiverse Computing

Input: $0.070

Output: $0.100

Pricing

Input (1M): $0.070
Output (1M): $0.100
Blended (3:1): $0.077

Performance

Tokens / s: 116.7
TTFT (token): 0.31s
TTFT (answer): 0.31s

Benchmarks

MMLU Pro 70.3%

GPQA 42.6%

HLE 4.3%

AIME —

LiveCodeBench 18.6%

SciCode 17.9%

Math 500 —

AA Indexes

Intelligence

22.1%

Coding

12.9%

Math

10.0%

GPT-5 (ChatGPT)

OpenAI

Input: $1.250

Output: $10.000

Pricing

Input (1M): $1.250
Output (1M): $10.000
Blended (3:1): $3.438

Performance

Tokens / s: 169.6
TTFT (token): 0.67s
TTFT (answer): 0.67s

Benchmarks

MMLU Pro 82.0%

GPQA 68.6%

HLE 5.8%

AIME —

LiveCodeBench 54.3%

SciCode 37.8%

Math 500 —

AA Indexes

Intelligence

21.8%

Coding

21.2%

Math

48.3%

Hermes 4 - Llama-3.1 405B (Reasoning)

Nous Research

Input: $1.000

Output: $3.000

Pricing

Input (1M): $1.000
Output (1M): $3.000
Blended (3:1): $1.500

Performance

Tokens / s: 33.8
TTFT (token): 0.72s
TTFT (answer): 59.97s

Benchmarks

MMLU Pro 82.9%

GPQA 72.7%

HLE 10.3%

AIME —

LiveCodeBench 68.6%

SciCode 25.2%

Math 500 —

AA Indexes

Intelligence

21.7%

Coding

16.0%

Math

69.7%

GPT-5 mini (minimal)

OpenAI

Input: $0.250

Output: $2.000

Pricing

Input (1M): $0.250
Output (1M): $2.000
Blended (3:1): $0.688

Performance

Tokens / s: 72.1
TTFT (token): 0.67s
TTFT (answer): 0.67s

Benchmarks

MMLU Pro 77.5%

GPQA 68.7%

HLE 5.0%

AIME —

LiveCodeBench 54.5%

SciCode 36.9%

Math 500 —

AA Indexes

Intelligence

21.6%

Coding

21.9%

Math

46.7%

gpt-oss-20B (low)

OpenAI

Input: $0.070

Output: $0.200

Pricing

Input (1M): $0.070
Output (1M): $0.200
Blended (3:1): $0.100

Performance

Tokens / s: 246.9
TTFT (token): 0.51s
TTFT (answer): 8.61s

Benchmarks

MMLU Pro 71.8%

GPQA 61.1%

HLE 5.1%

AIME —

LiveCodeBench 65.2%

SciCode 34.0%

Math 500 —

AA Indexes

Intelligence

21.2%

Coding

14.4%

Math

62.3%

Mistral Medium 3.1

Mistral

Input: $0.400

Output: $2.000

Pricing

Input (1M): $0.400
Output (1M): $2.000
Blended (3:1): $0.800

Performance

Tokens / s: 90.6
TTFT (token): 0.39s
TTFT (answer): 0.39s

Benchmarks

MMLU Pro 68.3%

GPQA 58.8%

HLE 4.4%

AIME —

LiveCodeBench 40.6%

SciCode 33.8%

Math 500 —

AA Indexes

Intelligence

21.2%

Coding

18.3%

Math

38.3%

Gemini 2.5 Flash (Non-reasoning)

Google

Input: $0.300

Output: $2.500

Pricing

Input (1M): $0.300
Output (1M): $2.500
Blended (3:1): $0.850

Performance

Tokens / s: 224.2
TTFT (token): 0.41s
TTFT (answer): 0.41s

Benchmarks

MMLU Pro 80.9%

GPQA 68.3%

HLE 5.1%

AIME 50.0%

LiveCodeBench 49.5%

SciCode 29.1%

Math 500 93.2%

AA Indexes

Intelligence

21.0%

Coding

17.8%

Math

60.3%

Qwen3 VL 235B A22B Instruct

Alibaba

Input: $0.700

Output: $2.800

Pricing

Input (1M): $0.700
Output (1M): $2.800
Blended (3:1): $1.225

Performance

Tokens / s: 35.3
TTFT (token): 1.09s
TTFT (answer): 1.09s

Benchmarks

MMLU Pro 82.3%

GPQA 71.2%

HLE 6.3%

AIME —

LiveCodeBench 59.4%

SciCode 35.9%

Math 500 —

AA Indexes

Intelligence

20.9%

Coding

16.5%

Math

70.7%

Ring-flash-2.0

InclusionAI

Input: $0.140

Output: $0.570

Pricing

Input (1M): $0.140
Output (1M): $0.570
Blended (3:1): $0.247

Performance

Tokens / s: 87.3
TTFT (token): 1.50s
TTFT (answer): 24.41s

Benchmarks

MMLU Pro 79.3%

GPQA 72.5%

HLE 8.9%

AIME —

LiveCodeBench 62.8%

SciCode 16.8%

Math 500 —

AA Indexes

Intelligence

20.6%

Coding

10.6%

Math

83.7%

Hermes 4 - Llama-3.1 70B (Reasoning)

Nous Research

Input: $0.130

Output: $0.400

Pricing

Input (1M): $0.130
Output (1M): $0.400
Blended (3:1): $0.198

Performance

Tokens / s: 81.0
TTFT (token): 0.57s
TTFT (answer): 25.27s

Benchmarks

MMLU Pro 81.1%

GPQA 69.9%

HLE 7.9%

AIME —

LiveCodeBench 65.3%

SciCode 34.1%

Math 500 —

AA Indexes

Intelligence

20.4%

Coding

14.4%

Math

68.7%

Qwen3 Coder 30B A3B Instruct

Alibaba

Input: $0.450

Output: $2.250

Pricing

Input (1M): $0.450
Output (1M): $2.250
Blended (3:1): $0.900

Performance

Tokens / s: 98.6
TTFT (token): 1.57s
TTFT (answer): 1.57s

Benchmarks

MMLU Pro 70.6%

GPQA 51.6%

HLE 4.0%

AIME 29.7%

LiveCodeBench 40.3%

SciCode 27.8%

Math 500 89.3%

AA Indexes

Intelligence

20.3%

Coding

19.4%

Math

29.0%

Qwen3 Next 80B A3B Instruct

Alibaba

Input: $0.500

Output: $2.000

Pricing

Input (1M): $0.500
Output (1M): $2.000
Blended (3:1): $0.875

Performance

Tokens / s: 145.9
TTFT (token): 1.23s
TTFT (answer): 1.23s

Benchmarks

MMLU Pro 81.9%

GPQA 73.8%

HLE 7.3%

AIME —

LiveCodeBench 68.4%

SciCode 30.7%

Math 500 —

AA Indexes

Intelligence

20.3%

Coding

15.3%

Math

66.3%

Codestral (Jan '25)

Mistral

Input: $0.300

Output: $0.900

Pricing

Input (1M): $0.300
Output (1M): $0.900
Blended (3:1): $0.450

Performance

Tokens / s: 212.0
TTFT (token): 0.39s
TTFT (answer): 0.39s

Benchmarks

MMLU Pro 44.6%

GPQA 31.2%

HLE 4.5%

AIME 4.3%

LiveCodeBench 24.3%

SciCode 24.7%

Math 500 60.7%

AA Indexes

Intelligence

20.1%

Coding

16.3%

Math

6.0%

Gemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning)

Google

Input: $0.100

Output: $0.400

Pricing

Input (1M): $0.100
Output (1M): $0.400
Blended (3:1): $0.175

Performance

Tokens / s: 450.6
TTFT (token): 0.34s
TTFT (answer): 0.34s

Benchmarks

MMLU Pro 79.6%

GPQA 65.1%

HLE 4.6%

AIME —

LiveCodeBench 64.1%

SciCode 28.5%

Math 500 —

AA Indexes

Intelligence

20.1%

Coding

14.5%

Math

46.7%

Qwen3 235B A22B (Reasoning)

Alibaba

Input: $0.700

Output: $8.400

Pricing

Input (1M): $0.700
Output (1M): $8.400
Blended (3:1): $2.625

Performance

Tokens / s: 54.2
TTFT (token): 1.22s
TTFT (answer): 38.14s

Benchmarks

MMLU Pro 82.8%

GPQA 70.0%

HLE 11.7%

AIME 84.0%

LiveCodeBench 62.2%

SciCode 39.9%

Math 500 93.0%

AA Indexes

Intelligence

20.0%

Coding

17.4%

Math

82.0%

Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)

NVIDIA

Input: $0.600

Output: $1.800

Pricing

Input (1M): $0.600
Output (1M): $1.800
Blended (3:1): $0.900

Performance

Tokens / s: 37.2
TTFT (token): 0.68s
TTFT (answer): 54.46s

Benchmarks

MMLU Pro 82.5%

GPQA 72.8%

HLE 8.1%

AIME 74.7%

LiveCodeBench 64.1%

SciCode 34.7%

Math 500 95.2%

AA Indexes

Intelligence

20.0%

Coding

13.1%

Math

63.7%

Ling-flash-2.0

InclusionAI

Input: $0.140

Output: $0.570

Pricing

Input (1M): $0.140
Output (1M): $0.570
Blended (3:1): $0.247

Performance

Tokens / s: 54.6
TTFT (token): 1.56s
TTFT (answer): 1.56s

Benchmarks

MMLU Pro 77.7%

GPQA 65.7%

HLE 6.3%

AIME —

LiveCodeBench 58.9%

SciCode 28.9%

Math 500 —

AA Indexes

Intelligence

19.9%

Coding

16.7%

Math

65.3%

Qwen3 VL 30B A3B (Reasoning)

Alibaba

Input: $0.200

Output: $2.400

Pricing

Input (1M): $0.200
Output (1M): $2.400
Blended (3:1): $0.750

Performance

Tokens / s: 106.6
TTFT (token): 0.97s
TTFT (answer): 19.74s

Benchmarks

MMLU Pro 80.7%

GPQA 72.0%

HLE 8.7%

AIME —

LiveCodeBench 69.7%

SciCode 28.8%

Math 500 —

AA Indexes

Intelligence

19.8%

Coding

13.1%

Math

82.3%

QwQ 32B

Alibaba

Input: $0.430

Output: $0.600

Pricing

Input (1M): $0.430
Output (1M): $0.600
Blended (3:1): $0.473

Performance

Tokens / s: 29.6
TTFT (token): 1.14s
TTFT (answer): 85.17s

Benchmarks

MMLU Pro 76.4%

GPQA 59.3%

HLE 8.2%

AIME 78.0%

LiveCodeBench 63.1%

SciCode 35.8%

Math 500 95.7%

AA Indexes

Intelligence

19.7%

Coding

—

Math

29.0%

Llama Nemotron Super 49B v1.5 (Reasoning)

NVIDIA

Input: $0.100

Output: $0.400

Pricing

Input (1M): $0.100
Output (1M): $0.400
Blended (3:1): $0.175

Performance

Tokens / s: 74.1
TTFT (token): 0.42s
TTFT (answer): 27.39s

Benchmarks

MMLU Pro 81.4%

GPQA 74.8%

HLE 6.8%

AIME 86.0%

LiveCodeBench 73.7%

SciCode 34.8%

Math 500 98.3%

AA Indexes

Intelligence

19.4%

Coding

15.2%

Math

76.7%

Reka Core

Reka AI

Input: $2.000

Output: $2.000

Pricing

Input (1M): $2.000
Output (1M): $2.000
Blended (3:1): $2.000

Performance

Tokens / s: 40.9
TTFT (token): 1.42s
TTFT (answer): 1.42s

Benchmarks

MMLU Pro —

GPQA —

HLE —

AIME —

LiveCodeBench —

SciCode —

Math 500 55.8%

AA Indexes

Intelligence

19.4%

Coding

—

Math

—

GLM-4.5V (Reasoning)

Z AI

Input: $0.600

Output: $1.800

Pricing

Input (1M): $0.600
Output (1M): $1.800
Blended (3:1): $0.900

Performance

Tokens / s: 31.4
TTFT (token): 0.77s
TTFT (answer): 64.44s

Benchmarks

MMLU Pro 78.8%

GPQA 68.4%

HLE 5.9%

AIME —

LiveCodeBench 60.4%

SciCode 22.1%

Math 500 —

AA Indexes

Intelligence

19.3%

Coding

10.9%

Math

73.0%

Nova Premier

Amazon

Input: $2.500

Output: $12.500

Pricing

Input (1M): $2.500
Output (1M): $12.500
Blended (3:1): $5.000

Performance

Tokens / s: 75.1
TTFT (token): 0.81s
TTFT (answer): 0.81s

Benchmarks

MMLU Pro 73.3%

GPQA 56.9%

HLE 4.7%

AIME 17.0%

LiveCodeBench 31.7%

SciCode 27.9%

Math 500 83.9%

AA Indexes

Intelligence

19.3%

Coding

13.8%

Math

17.3%

Magistral Medium 1

Mistral

Input: $2.000

Output: $5.000

Pricing

Input (1M): $2.000
Output (1M): $5.000
Blended (3:1): $2.750

Performance

Tokens / s: 60.2
TTFT (token): 0.49s
TTFT (answer): 33.71s

Benchmarks

MMLU Pro 75.3%

GPQA 67.9%

HLE 9.5%

AIME 70.0%

LiveCodeBench 52.7%

SciCode 29.7%

Math 500 91.7%

AA Indexes

Intelligence

19.1%

Coding

16.0%

Math

40.3%

GPT-4o (Aug '24)

OpenAI

Input: $2.500

Output: $10.000

Pricing

Input (1M): $2.500
Output (1M): $10.000
Blended (3:1): $4.375

Performance

Tokens / s: 79.0
TTFT (token): 0.53s
TTFT (answer): 0.53s

Benchmarks

MMLU Pro —

GPQA 52.1%

HLE 2.9%

AIME 11.7%

LiveCodeBench 31.7%

SciCode 33.1%

Math 500 79.5%

AA Indexes

Intelligence

19.0%

Coding

16.6%

Math

—

Llama 4 Maverick

Meta

Input: $0.310

Output: $0.850

Pricing

Input (1M): $0.310
Output (1M): $0.850
Blended (3:1): $0.461

Performance

Tokens / s: 127.7
TTFT (token): 0.46s
TTFT (answer): 0.46s

Benchmarks

MMLU Pro 80.9%

GPQA 67.1%

HLE 4.8%

AIME 39.0%

LiveCodeBench 39.7%

SciCode 33.1%

Math 500 88.9%

AA Indexes

Intelligence

19.0%

Coding

15.6%

Math

19.3%

Gemini 2.0 Flash (Feb '25)

Google

Input: $0.100

Output: $0.400

Pricing

Input (1M): $0.100
Output (1M): $0.400
Blended (3:1): $0.175

Performance

Tokens / s: —
TTFT (token): —
TTFT (answer): —

Benchmarks

MMLU Pro 77.9%

GPQA 62.3%

HLE 5.3%

AIME 33.0%

LiveCodeBench 33.4%

SciCode 33.3%

Math 500 93.0%

AA Indexes

Intelligence

19.0%

Coding

13.6%

Math

21.7%

Devstral Medium

Mistral

Input: $0.400

Output: $2.000

Pricing

Input (1M): $0.400
Output (1M): $2.000
Blended (3:1): $0.800

Performance

Tokens / s: 112.4
TTFT (token): 0.43s
TTFT (answer): 0.43s

Benchmarks

MMLU Pro 70.8%

GPQA 49.2%

HLE 3.8%

AIME 6.7%

LiveCodeBench 33.7%

SciCode 29.4%

Math 500 70.7%

AA Indexes

Intelligence

18.9%

Coding

15.9%

Math

4.7%

Nova 2.0 Lite (Non-reasoning)

Amazon

Input: $0.300

Output: $2.500

Pricing

Input (1M): $0.300
Output (1M): $2.500
Blended (3:1): $0.850

Performance

Tokens / s: 224.9
TTFT (token): 0.51s
TTFT (answer): 0.51s

Benchmarks

MMLU Pro 74.3%

GPQA 60.3%

HLE 3.0%

AIME —

LiveCodeBench 34.6%

SciCode 24.0%

Math 500 —

AA Indexes

Intelligence

18.9%

Coding

12.5%

Math

33.7%

Claude 3.5 Haiku

Anthropic

Input: $0.800

Output: $4.000

Pricing

Input (1M): $0.800
Output (1M): $4.000
Blended (3:1): $1.600

Performance

Tokens / s: 48.4
TTFT (token): 0.70s
TTFT (answer): 0.70s

Benchmarks

MMLU Pro 63.4%

GPQA 40.8%

HLE 3.5%

AIME 3.3%

LiveCodeBench 31.4%

SciCode 27.4%

Math 500 72.1%

AA Indexes

Intelligence

18.8%

Coding

10.7%

Math

—

DeepSeek R1 (Jan '25)

DeepSeek

Input: $1.350

Output: $4.000

Pricing

Input (1M): $1.350
Output (1M): $4.000
Blended (3:1): $2.362

Performance

Tokens / s: —
TTFT (token): —
TTFT (answer): —

Benchmarks

MMLU Pro 84.4%

GPQA 70.8%

HLE 9.3%

AIME 68.3%

LiveCodeBench 61.7%

SciCode 35.7%

Math 500 96.6%

AA Indexes

Intelligence

18.8%

Coding

15.9%

Math

68.0%

Reka Flash (Feb '24)

Reka AI

Input: $0.200

Output: $0.800

Pricing

Input (1M): $0.200
Output (1M): $0.800
Blended (3:1): $0.350

Performance

Tokens / s: 69.1
TTFT (token): 1.32s
TTFT (answer): 1.32s

Benchmarks

MMLU Pro —

GPQA —

HLE —

AIME —

LiveCodeBench —

SciCode —

Math 500 32.6%

AA Indexes

Intelligence

18.7%

Coding

—

Math

—

GPT-4o (March 2025, chatgpt-4o-latest)

OpenAI

Input: $5.000

Output: $15.000

Pricing

Input (1M): $5.000
Output (1M): $15.000
Blended (3:1): $7.500

Performance

Tokens / s: 215.5
TTFT (token): 0.51s
TTFT (answer): 0.51s

Benchmarks

MMLU Pro 80.3%

GPQA 65.5%

HLE 5.0%

AIME 32.7%

LiveCodeBench 42.5%

SciCode 36.6%

Math 500 89.3%

AA Indexes

Intelligence

18.6%

Coding

—

Math

25.7%

Reka Edge

Reka AI

Input: $0.100

Output: $0.100

Pricing

Input (1M): $0.100
Output (1M): $0.100
Blended (3:1): $0.100

Performance

Tokens / s: 63.8
TTFT (token): 1.29s
TTFT (answer): 1.29s

Benchmarks

MMLU Pro —

GPQA —

HLE —

AIME —

LiveCodeBench —

SciCode —

Math 500 21.6%

AA Indexes

Intelligence

18.5%

Coding

—

Math

—

Gemini 2.5 Flash-Lite (Reasoning)

Google

Input: $0.100

Output: $0.400

Pricing

Input (1M): $0.100
Output (1M): $0.400
Blended (3:1): $0.175

Performance

Tokens / s: 369.9
TTFT (token): 15.73s
TTFT (answer): 15.73s

Benchmarks

MMLU Pro 75.9%

GPQA 62.5%

HLE 6.4%

AIME 70.3%

LiveCodeBench 59.3%

SciCode 19.3%

Math 500 96.9%

AA Indexes

Intelligence

18.0%

Coding

9.5%

Math

53.3%

Sonar Reasoning

Perplexity

Input: $1.000

Output: $5.000

Pricing

Input (1M): $1.000
Output (1M): $5.000
Blended (3:1): $2.000

Performance

Tokens / s: —
TTFT (token): —
TTFT (answer): —

Benchmarks

MMLU Pro —

GPQA 62.3%

HLE —

AIME 77.0%

LiveCodeBench —

SciCode —

Math 500 92.1%

AA Indexes

Intelligence

17.9%

Coding

—

Math

—

Devstral Small (May '25)

Mistral

Input: $0.100

Output: $0.300

Pricing

Input (1M): $0.100
Output (1M): $0.300
Blended (3:1): $0.150

Performance

Tokens / s: 200.0
TTFT (token): 0.36s
TTFT (answer): 0.36s

Benchmarks

MMLU Pro 63.2%

GPQA 43.4%

HLE 4.0%

AIME 6.7%

LiveCodeBench 25.8%

SciCode 24.5%

Math 500 68.4%

AA Indexes

Intelligence

17.9%

Coding

12.2%

Math

—

Llama 3.1 Instruct 405B

Meta

Input: $3.750

Output: $6.750

Pricing

Input (1M): $3.750
Output (1M): $6.750
Blended (3:1): $4.188

Performance

Tokens / s: 24.9
TTFT (token): 0.77s
TTFT (answer): 0.77s

Benchmarks

MMLU Pro 73.2%

GPQA 51.5%

HLE 4.2%

AIME 21.3%

LiveCodeBench 30.5%

SciCode 29.9%

Math 500 70.3%

AA Indexes

Intelligence

17.6%

Coding

14.5%

Math

3.0%

Mistral Medium 3

Mistral

Input: $0.400

Output: $2.000

Pricing

Input (1M): $0.400
Output (1M): $2.000
Blended (3:1): $0.800

Performance

Tokens / s: 95.7
TTFT (token): 0.38s
TTFT (answer): 0.38s

Benchmarks

MMLU Pro 76.0%

GPQA 57.8%

HLE 4.3%

AIME 44.0%

LiveCodeBench 40.0%

SciCode 33.1%

Math 500 90.7%

AA Indexes

Intelligence

17.6%

Coding

13.6%

Math

30.3%

GLM-4.6V (Non-reasoning)

Z AI

Input: $0.300

Output: $0.900

Pricing

Input (1M): $0.300
Output (1M): $0.900
Blended (3:1): $0.450

Performance

Tokens / s: 46.6
TTFT (token): 1.01s
TTFT (answer): 1.01s

Benchmarks

MMLU Pro 75.2%

GPQA 56.6%

HLE 3.7%

AIME —

LiveCodeBench 41.1%

SciCode 27.2%

Math 500 —

AA Indexes

Intelligence

17.4%

Coding

11.1%

Math

26.3%

Nova 2.0 Omni (Non-reasoning)

Amazon

Input: $0.300

Output: $2.500

Pricing

Input (1M): $0.300
Output (1M): $2.500
Blended (3:1): $0.850

Performance

Tokens / s: 226.3
TTFT (token): 0.72s
TTFT (answer): 0.72s

Benchmarks

MMLU Pro 71.9%

GPQA 55.5%

HLE 3.9%

AIME —

LiveCodeBench 30.5%

SciCode 27.9%

Math 500 —

AA Indexes

Intelligence

17.3%

Coding

13.8%

Math

37.0%

Qwen3 235B A22B (Non-reasoning)

Alibaba

Input: $0.700

Output: $2.800

Pricing

Input (1M): $0.700
Output (1M): $2.800
Blended (3:1): $1.225

Performance

Tokens / s: 45.7
TTFT (token): 1.10s
TTFT (answer): 1.10s

Benchmarks

MMLU Pro 76.2%

GPQA 61.3%

HLE 4.7%

AIME 32.7%

LiveCodeBench 34.3%

SciCode 29.9%

Math 500 90.2%

AA Indexes

Intelligence

17.3%

Coding

14.0%

Math

23.7%

ERNIE 4.5 300B A47B

Baidu

Input: $0.280

Output: $1.100

Pricing

Input (1M): $0.280
Output (1M): $1.100
Blended (3:1): $0.485

Performance

Tokens / s: 26.4
TTFT (token): 2.03s
TTFT (answer): 2.03s

Benchmarks

MMLU Pro 77.6%

GPQA 81.1%

HLE 3.5%

AIME 49.3%

LiveCodeBench 46.7%

SciCode 31.5%

Math 500 93.1%

AA Indexes

Intelligence

17.3%

Coding

14.5%

Math

41.3%

Qwen3 VL 32B Instruct

Alibaba

Input: $0.700

Output: $2.800

Pricing

Input (1M): $0.700
Output (1M): $2.800
Blended (3:1): $1.225

Performance

Tokens / s: 45.6
TTFT (token): 0.95s
TTFT (answer): 0.95s

Benchmarks

MMLU Pro 79.1%

GPQA 67.1%

HLE 6.3%

AIME —

LiveCodeBench 51.4%

SciCode 30.1%

Math 500 —

AA Indexes

Intelligence

17.2%

Coding

15.6%

Math

68.3%

DeepSeek R1 Distill Qwen 32B

DeepSeek

Input: $0.285

Output: $0.285

Pricing

Input (1M): $0.285
Output (1M): $0.285
Blended (3:1): $0.285

Performance

Tokens / s: 39.4
TTFT (token): 0.38s
TTFT (answer): 51.14s

Benchmarks

MMLU Pro 73.9%

GPQA 61.5%

HLE 5.5%

AIME 68.7%

LiveCodeBench 27.0%

SciCode 37.6%

Math 500 94.1%

AA Indexes

Intelligence

17.2%

Coding

—

Math

63.0%

Qwen3 VL 8B (Reasoning)

Alibaba

Input: $0.180

Output: $2.100

Pricing

Input (1M): $0.180
Output (1M): $2.100
Blended (3:1): $0.660

Performance

Tokens / s: 63.6
TTFT (token): 0.93s
TTFT (answer): 32.39s

Benchmarks

MMLU Pro 74.9%

GPQA 57.9%

HLE 3.3%

AIME —

LiveCodeBench 35.3%

SciCode 21.9%

Math 500 —

AA Indexes

Intelligence

17.1%

Coding

9.8%

Math

30.7%

Qwen3 32B (Reasoning)

Alibaba

Input: $0.700

Output: $8.400

Pricing

Input (1M): $0.700
Output (1M): $8.400
Blended (3:1): $2.625

Performance

Tokens / s: 88.7
TTFT (token): 1.03s
TTFT (answer): 23.57s

Benchmarks

MMLU Pro 79.8%

GPQA 66.8%

HLE 8.3%

AIME 80.7%

LiveCodeBench 54.6%

SciCode 35.4%

Math 500 96.1%

AA Indexes

Intelligence

17.1%

Coding

13.8%

Math

73.0%

DeepSeek V3 (Dec '24)

DeepSeek

Input: $0.400

Output: $0.890

Pricing

Input (1M): $0.400
Output (1M): $0.890
Blended (3:1): $0.625

Performance

Tokens / s: —
TTFT (token): —
TTFT (answer): —

Benchmarks

MMLU Pro 75.2%

GPQA 55.7%

HLE 3.6%

AIME 25.3%

LiveCodeBench 35.9%

SciCode 35.4%

Math 500 88.7%

AA Indexes

Intelligence

17.1%

Coding

16.4%

Math

26.0%

Hermes 4 - Llama-3.1 405B (Non-reasoning)

Nous Research

Input: $1.000

Output: $3.000

Pricing

Input (1M): $1.000
Output (1M): $3.000
Blended (3:1): $1.500

Performance

Tokens / s: 32.9
TTFT (token): 0.72s
TTFT (answer): 0.72s

Benchmarks

MMLU Pro 72.9%

GPQA 53.6%

HLE 4.2%

AIME —

LiveCodeBench 54.6%

SciCode 34.6%

Math 500 —

AA Indexes

Intelligence

17.1%

Coding

18.1%

Math

15.3%

EXAONE 4.0 32B (Reasoning)

LG AI Research

Input: $0.600

Output: $1.000

Pricing

Input (1M): $0.600
Output (1M): $1.000
Blended (3:1): $0.700

Performance

Tokens / s: 96.2
TTFT (token): 0.30s
TTFT (answer): 21.08s

Benchmarks

MMLU Pro 81.8%

GPQA 73.9%

HLE 10.5%

AIME 84.3%

LiveCodeBench 74.7%

SciCode 34.4%

Math 500 97.7%

AA Indexes

Intelligence

17.0%

Coding

14.0%

Math

80.0%

Qwen3 14B (Reasoning)

Alibaba

Input: $0.350

Output: $4.200

Pricing

Input (1M): $0.350
Output (1M): $4.200
Blended (3:1): $1.313

Performance

Tokens / s: 58.8
TTFT (token): 1.04s
TTFT (answer): 35.08s

Benchmarks

MMLU Pro 77.4%

GPQA 60.4%

HLE 4.3%

AIME 76.3%

LiveCodeBench 52.3%

SciCode 31.6%

Math 500 96.1%

AA Indexes

Intelligence

16.8%

Coding

13.1%

Math

55.7%

Magistral Small 1

Mistral

Input: $0.500

Output: $1.500

Pricing

Input (1M): $0.500
Output (1M): $1.500
Blended (3:1): $0.750

Performance

Tokens / s: 118.7
TTFT (token): 0.35s
TTFT (answer): 17.20s

Benchmarks

MMLU Pro 74.6%

GPQA 64.1%

HLE 7.2%

AIME 71.3%

LiveCodeBench 51.4%

SciCode 24.1%

Math 500 96.3%

AA Indexes

Intelligence

16.8%

Coding

11.1%

Math

41.3%

Olmo 3 7B Think

Allen Institute for AI

Input: $0.120

Output: $0.200

Pricing

Input (1M): $0.120
Output (1M): $0.200
Blended (3:1): $0.140

Performance

Tokens / s: 170.3
TTFT (token): 0.36s
TTFT (answer): 12.11s

Benchmarks

MMLU Pro 65.5%

GPQA 51.6%

HLE 5.7%

AIME —

LiveCodeBench 61.7%

SciCode 21.2%

Math 500 —

AA Indexes

Intelligence

16.8%

Coding

7.6%

Math

70.7%

CompactifAI Llama 3.3 70B Slim

Multiverse Computing

Input: $0.160

Output: $0.310

Pricing

Input (1M): $0.160
Output (1M): $0.310
Blended (3:1): $0.198

Performance

Tokens / s: 130.7
TTFT (token): 0.30s
TTFT (answer): 0.30s

Benchmarks

MMLU Pro 57.1%

GPQA 35.5%

HLE 2.6%

AIME —

LiveCodeBench 21.0%

SciCode 3.2%

Math 500 —

AA Indexes

Intelligence

16.5%

Coding

8.1%

Math

3.3%

CompactifAI Mistral Small 3.1 Slim

Multiverse Computing

Input: $0.050

Output: $0.080

Pricing

Input (1M): $0.050
Output (1M): $0.080
Blended (3:1): $0.058

Performance

Tokens / s: 121.6
TTFT (token): 0.31s
TTFT (answer): 0.31s

Benchmarks

MMLU Pro 53.8%

GPQA 32.9%

HLE 5.1%

AIME —

LiveCodeBench 16.8%

SciCode 12.8%

Math 500 —

AA Indexes

Intelligence

16.5%

Coding

10.1%

Math

100.0%

Qwen3 VL 30B A3B Instruct

Alibaba

Input: $0.200

Output: $0.800

Pricing

Input (1M): $0.200
Output (1M): $0.800
Blended (3:1): $0.350

Performance

Tokens / s: 99.2
TTFT (token): 0.90s
TTFT (answer): 0.90s

Benchmarks

MMLU Pro 76.4%

GPQA 69.5%

HLE 6.4%

AIME —

LiveCodeBench 47.6%

SciCode 30.8%

Math 500 —

AA Indexes

Intelligence

16.4%

Coding

14.3%

Math

72.3%

DeepSeek R1 0528 Qwen3 8B

DeepSeek

Input: $0.060

Output: $0.090

Pricing

Input (1M): $0.060
Output (1M): $0.090
Blended (3:1): $0.068

Performance

Tokens / s: —
TTFT (token): —
TTFT (answer): —

Benchmarks

MMLU Pro 73.9%

GPQA 61.2%

HLE 5.6%

AIME 65.0%

LiveCodeBench 51.3%

SciCode 20.4%

Math 500 93.2%

AA Indexes

Intelligence

16.4%

Coding

7.8%

Math

63.7%

Ministral 3 14B

Mistral

Input: $0.200

Output: $0.200

Pricing

Input (1M): $0.200
Output (1M): $0.200
Blended (3:1): $0.200

Performance

Tokens / s: 126.2
TTFT (token): 0.30s
TTFT (answer): 0.30s

Benchmarks

MMLU Pro 69.3%

GPQA 57.2%

HLE 4.6%

AIME —

LiveCodeBench 35.1%

SciCode 23.6%

Math 500 —

AA Indexes

Intelligence

16.3%

Coding

10.9%

Math

30.0%

Qwen2.5 Max

Alibaba

Input: $1.600

Output: $6.400

Pricing

Input (1M): $1.600
Output (1M): $6.400
Blended (3:1): $2.800

Performance

Tokens / s: 41.4
TTFT (token): 1.13s
TTFT (answer): 1.13s

Benchmarks

MMLU Pro 76.2%

GPQA 58.7%

HLE 4.5%

AIME 23.3%

LiveCodeBench 35.9%

SciCode 33.7%

Math 500 83.5%

AA Indexes

Intelligence

16.3%

Coding

—

Math

—

DeepSeek R1 Distill Llama 70B

DeepSeek

Input: $0.875

Output: $1.300

Pricing

Input (1M): $0.875
Output (1M): $1.300
Blended (3:1): $0.963

Performance

Tokens / s: 38.9
TTFT (token): 0.95s
TTFT (answer): 52.36s

Benchmarks

MMLU Pro 79.5%

GPQA 40.2%

HLE 6.1%

AIME 67.0%

LiveCodeBench 26.6%

SciCode 31.2%

Math 500 93.5%

AA Indexes

Intelligence

16.0%

Coding

11.4%

Math

53.7%

Claude 3.5 Sonnet (Oct '24)

Anthropic

Input: $3.000

Output: $15.000

Pricing

Input (1M): $3.000
Output (1M): $15.000
Blended (3:1): $6.000

Performance

Tokens / s: —
TTFT (token): —
TTFT (answer): —

Benchmarks

MMLU Pro 77.2%

GPQA 59.9%

HLE 3.9%

AIME 15.7%

LiveCodeBench 38.1%

SciCode 36.6%

Math 500 77.1%

AA Indexes

Intelligence

15.9%

Coding

30.2%

Math

—

DeepSeek R1 Distill Qwen 14B

DeepSeek

Input: $0.150

Output: $0.150

Pricing

Input (1M): $0.150
Output (1M): $0.150
Blended (3:1): $0.150

Performance

Tokens / s: —
TTFT (token): —
TTFT (answer): —

Benchmarks

MMLU Pro 74.0%

GPQA 48.4%

HLE 4.4%

AIME 66.7%

LiveCodeBench 37.6%

SciCode 23.9%

Math 500 94.9%

AA Indexes

Intelligence

15.8%

Coding

—

Math

55.7%

Qwen3 30B A3B (Reasoning)

Alibaba

Input: $0.200

Output: $2.400

Pricing

Input (1M): $0.200
Output (1M): $2.400
Blended (3:1): $0.750

Performance

Tokens / s: 66.5
TTFT (token): 1.09s
TTFT (answer): 31.19s

Benchmarks

MMLU Pro 77.7%

GPQA 61.6%

HLE 6.6%

AIME 75.3%

LiveCodeBench 50.6%

SciCode 28.5%

Math 500 95.9%

AA Indexes

Intelligence

15.8%

Coding

11.0%

Math

72.3%

Qwen3 Omni 30B A3B (Reasoning)

Alibaba

Input: $0.250

Output: $0.970

Pricing

Input (1M): $0.250
Output (1M): $0.970
Blended (3:1): $0.430

Performance

Tokens / s: 97.1
TTFT (token): 0.93s
TTFT (answer): 21.52s

Benchmarks

MMLU Pro 79.2%

GPQA 72.6%

HLE 7.3%

AIME —

LiveCodeBench 67.9%

SciCode 30.6%

Math 500 —

AA Indexes

Intelligence

15.8%

Coding

12.7%

Math

74.0%

Devstral Small (Jul '25)

Mistral

Input: $0.100

Output: $0.300

Pricing

Input (1M): $0.100
Output (1M): $0.300
Blended (3:1): $0.150

Performance

Tokens / s: 237.5
TTFT (token): 0.37s
TTFT (answer): 0.37s

Benchmarks

MMLU Pro 62.2%

GPQA 41.4%

HLE 3.7%

AIME 0.3%

LiveCodeBench 25.4%

SciCode 24.3%

Math 500 63.5%

AA Indexes

Intelligence

15.7%

Coding

12.1%

Math

29.3%

Qwen3 30B A3B 2507 Instruct

Alibaba

Input: $0.200

Output: $0.800

Pricing

Input (1M): $0.200
Output (1M): $0.800
Blended (3:1): $0.350

Performance

Tokens / s: 59.8
TTFT (token): 1.00s
TTFT (answer): 1.00s

Benchmarks

MMLU Pro 77.7%

GPQA 65.9%

HLE 6.8%

AIME 72.7%

LiveCodeBench 51.5%

SciCode 30.4%

Math 500 97.5%

AA Indexes

Intelligence

15.5%

Coding

14.2%

Math

66.3%

Llama Nemotron Super 49B v1.5 (Non-reasoning)

NVIDIA

Input: $0.100

Output: $0.400

Pricing

Input (1M): $0.100
Output (1M): $0.400
Blended (3:1): $0.175

Performance

Tokens / s: 69.4
TTFT (token): 0.42s
TTFT (answer): 0.42s

Benchmarks

MMLU Pro 69.2%

GPQA 48.1%

HLE 4.3%

AIME 13.7%

LiveCodeBench 29.0%

SciCode 23.8%

Math 500 77.0%

AA Indexes

Intelligence

15.5%

Coding

10.5%

Math

8.0%

NVIDIA Nemotron Nano 9B V2 (Reasoning)

NVIDIA

Input: $0.040

Output: $0.160

Pricing

Input (1M): $0.040
Output (1M): $0.160
Blended (3:1): $0.070

Performance

Tokens / s: 118.2
TTFT (token): 0.39s
TTFT (answer): 17.31s

Benchmarks

MMLU Pro 74.2%

GPQA 57.0%

HLE 4.6%

AIME —

LiveCodeBench 72.4%

SciCode 22.0%

Math 500 —

AA Indexes

Intelligence

15.5%

Coding

8.3%

Math

69.7%

Sonar

Perplexity

Input: $1.000

Output: $1.000

Pricing

Input (1M): $1.000
Output (1M): $1.000
Blended (3:1): $1.000

Performance

Tokens / s: 78.3
TTFT (token): 1.56s
TTFT (answer): 1.56s

Benchmarks

MMLU Pro 68.9%

GPQA 47.1%

HLE 7.3%

AIME 48.7%

LiveCodeBench 29.5%

SciCode 22.9%

Math 500 81.7%

AA Indexes

Intelligence

15.5%

Coding

—

Math

—

Mistral Small 3.2

Mistral

Input: $0.100

Output: $0.300

Pricing

Input (1M): $0.100
Output (1M): $0.300
Blended (3:1): $0.150

Performance

Tokens / s: 111.3
TTFT (token): 0.29s
TTFT (answer): 0.29s

Benchmarks

MMLU Pro 68.1%

GPQA 50.5%

HLE 4.3%

AIME 32.3%

LiveCodeBench 27.5%

SciCode 26.4%

Math 500 88.3%

AA Indexes

Intelligence

15.4%

Coding

13.3%

Math

27.0%

Qwen3 8B (Reasoning)

Alibaba

Input: $0.180

Output: $2.100

Pricing

Input (1M): $0.180
Output (1M): $2.100
Blended (3:1): $0.660

Performance

Tokens / s: 85.1
TTFT (token): 0.95s
TTFT (answer): 24.46s

Benchmarks

MMLU Pro 74.3%

GPQA 58.9%

HLE 4.2%

AIME 74.7%

LiveCodeBench 40.6%

SciCode 22.6%

Math 500 90.4%

AA Indexes

Intelligence

15.3%

Coding

9.0%

Math

19.0%

Ministral 3 8B

Mistral

Input: $0.150

Output: $0.150

Pricing

Input (1M): $0.150
Output (1M): $0.150
Blended (3:1): $0.150

Performance

Tokens / s: 196.9
TTFT (token): 0.28s
TTFT (answer): 0.28s

Benchmarks

MMLU Pro 64.2%

GPQA 47.1%

HLE 4.3%

AIME —

LiveCodeBench 30.3%

SciCode 20.8%

Math 500 —

AA Indexes

Intelligence

15.2%

Coding

10.0%

Math

31.7%

NVIDIA Nemotron Nano 12B v2 VL (Reasoning)

NVIDIA

Input: $0.200

Output: $0.600

Pricing

Input (1M): $0.200
Output (1M): $0.600
Blended (3:1): $0.300

Performance

Tokens / s: 128.9
TTFT (token): 0.21s
TTFT (answer): 15.72s

Benchmarks

MMLU Pro 75.9%

GPQA 57.2%

HLE 5.3%

AIME —

LiveCodeBench 69.4%

SciCode 26.2%

Math 500 —

AA Indexes

Intelligence

15.2%

Coding

11.8%

Math

75.0%

QwQ 32B-Preview

Alibaba

Input: $0.120

Output: $0.180

Pricing

Input (1M): $0.120
Output (1M): $0.180
Blended (3:1): $0.135

Performance

Tokens / s: 40.8
TTFT (token): 0.42s
TTFT (answer): 49.44s

Benchmarks

MMLU Pro 64.8%

GPQA 55.7%

HLE 4.8%

AIME 45.3%

LiveCodeBench 33.7%

SciCode 3.8%

Math 500 91.0%

AA Indexes

Intelligence

15.2%

Coding

—

Math

—

Sonar Pro

Perplexity

Input: $3.000

Output: $15.000

Pricing

Input (1M): $3.000
Output (1M): $15.000
Blended (3:1): $6.000

Performance

Tokens / s: 86.2
TTFT (token): 1.56s
TTFT (answer): 1.56s

Benchmarks

MMLU Pro 75.5%

GPQA 57.8%

HLE 7.9%

AIME 29.0%

LiveCodeBench 27.5%

SciCode 22.6%

Math 500 74.5%

AA Indexes

Intelligence

15.2%

Coding

—

Math

—

Llama 3.3 Instruct 70B

Meta

Input: $0.585

Output: $0.715

Pricing

Input (1M): $0.585
Output (1M): $0.715
Blended (3:1): $0.675

Performance

Tokens / s: 105.5
TTFT (token): 0.49s
TTFT (answer): 0.49s

Benchmarks

MMLU Pro 71.3%

GPQA 49.8%

HLE 4.0%

AIME 30.0%

LiveCodeBench 28.8%

SciCode 26.0%

Math 500 77.3%

AA Indexes

Intelligence

15.1%

Coding

10.7%

Math

7.7%

Ling-mini-2.0

InclusionAI

Input: $0.070

Output: $0.280

Pricing

Input (1M): $0.070
Output (1M): $0.280
Blended (3:1): $0.122

Performance

Tokens / s: 156.9
TTFT (token): 2.12s
TTFT (answer): 2.12s

Benchmarks

MMLU Pro 67.1%

GPQA 56.2%

HLE 5.0%

AIME —

LiveCodeBench 42.9%

SciCode 13.5%

Math 500 —

AA Indexes

Intelligence

15.1%

Coding

5.0%

Math

49.3%

GPT-4o (Nov '24)

OpenAI

Input: $2.500

Output: $10.000

Pricing

Input (1M): $2.500
Output (1M): $10.000
Blended (3:1): $4.375

Performance

Tokens / s: 116.4
TTFT (token): 0.46s
TTFT (answer): 0.46s

Benchmarks

MMLU Pro 74.8%

GPQA 54.3%

HLE 3.3%

AIME 15.0%

LiveCodeBench 30.9%

SciCode 33.3%

Math 500 75.9%

AA Indexes

Intelligence

14.8%

Coding

16.7%

Math

6.0%

Qwen3 VL 8B Instruct

Alibaba

Input: $0.180

Output: $0.700

Pricing

Input (1M): $0.180
Output (1M): $0.700
Blended (3:1): $0.310

Performance

Tokens / s: 98.3
TTFT (token): 0.90s
TTFT (answer): 0.90s

Benchmarks

MMLU Pro 68.6%

GPQA 42.7%

HLE 2.9%

AIME —

LiveCodeBench 33.2%

SciCode 17.4%

Math 500 —

AA Indexes

Intelligence

14.7%

Coding

7.3%

Math

27.3%

Gemini 2.0 Flash-Lite (Feb '25)

Google

Input: $0.075

Output: $0.300

Pricing

Input (1M): $0.075
Output (1M): $0.300
Blended (3:1): $0.131

Performance

Tokens / s: —
TTFT (token): —
TTFT (answer): —

Benchmarks

MMLU Pro 72.4%

GPQA 53.5%

HLE 3.6%

AIME 27.7%

LiveCodeBench 18.5%

SciCode 25.0%

Math 500 87.3%

AA Indexes

Intelligence

14.7%

Coding

—

Math

—

Mistral Large 2 (Nov '24)

Mistral

Input: $2.000

Output: $6.000

Pricing

Input (1M): $2.000
Output (1M): $6.000
Blended (3:1): $3.000

Performance

Tokens / s: 45.9
TTFT (token): 0.44s
TTFT (answer): 0.44s

Benchmarks

MMLU Pro 69.7%

GPQA 48.6%

HLE 4.0%

AIME 11.0%

LiveCodeBench 29.3%

SciCode 29.2%

Math 500 73.6%

AA Indexes

Intelligence

14.7%

Coding

13.8%

Math

14.0%

Gemini 2.0 Flash-Lite (Preview)

Google

Input: $0.075

Output: $0.300

Pricing

Input (1M): $0.075
Output (1M): $0.300
Blended (3:1): $0.131

Performance

Tokens / s: —
TTFT (token): —
TTFT (answer): —

Benchmarks

MMLU Pro —

GPQA 54.2%

HLE 4.4%

AIME 30.3%

LiveCodeBench 17.9%

SciCode 24.7%

Math 500 87.3%

AA Indexes

Intelligence

14.5%

Coding

—

Math

—

GPT-4o (May '24)

OpenAI

Input: $5.000

Output: $15.000

Pricing

Input (1M): $5.000
Output (1M): $15.000
Blended (3:1): $7.500

Performance

Tokens / s: 78.4
TTFT (token): 0.51s
TTFT (answer): 0.51s

Benchmarks

MMLU Pro 74.0%

GPQA 52.6%

HLE 2.8%

AIME 11.0%

LiveCodeBench 33.4%

SciCode 30.9%

Math 500 79.1%

AA Indexes

Intelligence

14.5%

Coding

24.2%

Math

—

Qwen3 32B (Non-reasoning)

Alibaba

Input: $0.700

Output: $2.800

Pricing

Input (1M): $0.700
Output (1M): $2.800
Blended (3:1): $1.225

Performance

Tokens / s: 85.6
TTFT (token): 0.98s
TTFT (answer): 0.98s

Benchmarks

MMLU Pro 72.7%

GPQA 53.5%

HLE 4.3%

AIME 30.3%

LiveCodeBench 28.8%

SciCode 28.0%

Math 500 86.9%

AA Indexes

Intelligence

14.5%

Coding

—

Math

19.7%

Phi-3 Medium Instruct 14B

Microsoft Azure

Input: $0.170

Output: $0.680

Pricing

Input (1M): $0.170
Output (1M): $0.680
Blended (3:1): $0.297

Performance

Tokens / s: —
TTFT (token): —
TTFT (answer): —

Benchmarks

MMLU Pro 54.3%

GPQA 32.6%

HLE 4.5%

AIME 1.3%

LiveCodeBench 15.0%

SciCode 11.8%

Math 500 46.3%

AA Indexes

Intelligence

14.4%

Coding

8.9%

Math

1.3%

Reka Flash 3

Reka AI

Input: $0.200

Output: $0.800

Pricing

Input (1M): $0.200
Output (1M): $0.800
Blended (3:1): $0.350

Performance

Tokens / s: 49.3
TTFT (token): 1.31s
TTFT (answer): 41.91s

Benchmarks

MMLU Pro 66.9%

GPQA 52.9%

HLE 5.1%

AIME 51.0%

LiveCodeBench 43.5%

SciCode 26.7%

Math 500 89.3%

AA Indexes

Intelligence

14.3%

Coding

8.9%

Math

33.7%

Claude 3.5 Sonnet (June '24)

Anthropic

Input: $3.000

Output: $15.000

Pricing

Input (1M): $3.000
Output (1M): $15.000
Blended (3:1): $6.000

Performance

Tokens / s: —
TTFT (token): —
TTFT (answer): —

Benchmarks

MMLU Pro 75.1%

GPQA 56.0%

HLE 3.7%

AIME 9.7%

LiveCodeBench —

SciCode 31.6%

Math 500 69.5%

AA Indexes

Intelligence

14.2%

Coding

26.0%

Math

—

Qwen3 4B (Reasoning)

Alibaba

Input: $0.110

Output: $1.260

Pricing

Input (1M): $0.110
Output (1M): $1.260
Blended (3:1): $0.398

Performance

Tokens / s: 91.0
TTFT (token): 0.96s
TTFT (answer): 22.94s

Benchmarks

MMLU Pro 69.6%

GPQA 52.2%

HLE 5.1%

AIME 65.7%

LiveCodeBench 46.5%

SciCode 3.5%

Math 500 93.3%

AA Indexes

Intelligence

14.2%

Coding

—

Math

22.3%

Llama 3.1 Nemotron Instruct 70B

NVIDIA

Input: $1.200

Output: $1.200

Pricing

Input (1M): $1.200
Output (1M): $1.200
Blended (3:1): $1.200

Performance

Tokens / s: 39.6
TTFT (token): 0.34s
TTFT (answer): 0.34s

Benchmarks

MMLU Pro 69.0%

GPQA 46.5%

HLE 4.6%

AIME 24.7%

LiveCodeBench 16.9%

SciCode 23.3%

Math 500 73.3%

AA Indexes

Intelligence

14.1%

Coding

10.8%

Math

11.0%

GPT-4o (ChatGPT)

OpenAI

Input: $5.000

Output: $15.000

Pricing

Input (1M): $5.000
Output (1M): $15.000
Blended (3:1): $7.500

Performance

Tokens / s: 202.7
TTFT (token): 0.54s
TTFT (answer): 0.54s

Benchmarks

MMLU Pro 77.3%

GPQA 51.1%

HLE 3.7%

AIME 10.3%

LiveCodeBench —

SciCode 33.4%

Math 500 79.7%

AA Indexes

Intelligence

14.1%

Coding

—

Math

—

GPT-5 nano (minimal)

OpenAI

Input: $0.050

Output: $0.400

Pricing

Input (1M): $0.050
Output (1M): $0.400
Blended (3:1): $0.138

Performance

Tokens / s: 116.8
TTFT (token): 0.64s
TTFT (answer): 0.64s

Benchmarks

MMLU Pro 55.6%

GPQA 42.8%

HLE 4.1%

AIME —

LiveCodeBench 47.0%

SciCode 29.1%

Math 500 —

AA Indexes

Intelligence

14.1%

Coding

14.2%

Math

27.3%

Llama 4 Scout

Meta

Input: $0.180

Output: $0.625

Pricing

Input (1M): $0.180
Output (1M): $0.625
Blended (3:1): $0.287

Performance

Tokens / s: 141.7
TTFT (token): 0.51s
TTFT (answer): 0.51s

Benchmarks

MMLU Pro 75.2%

GPQA 58.7%

HLE 4.3%

AIME 28.3%

LiveCodeBench 29.9%

SciCode 17.0%

Math 500 84.4%

AA Indexes

Intelligence

14.0%

Coding

6.7%

Math

14.0%

NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning)

NVIDIA

Input: $0.060

Output: $0.240

Pricing

Input (1M): $0.060
Output (1M): $0.240
Blended (3:1): $0.105

Performance

Tokens / s: 174.9
TTFT (token): 0.23s
TTFT (answer): 0.23s

Benchmarks

MMLU Pro 57.9%

GPQA 39.9%

HLE 4.6%

AIME —

LiveCodeBench 36.0%

SciCode 23.0%

Math 500 —

AA Indexes

Intelligence

14.0%

Coding

15.8%

Math

13.3%

Mistral Small 3.1

Mistral

Input: $0.100

Output: $0.300

Pricing

Input (1M): $0.100
Output (1M): $0.300
Blended (3:1): $0.150

Performance

Tokens / s: 106.0
TTFT (token): 0.29s
TTFT (answer): 0.29s

Benchmarks

MMLU Pro 65.9%

GPQA 45.4%

HLE 4.8%

AIME 9.3%

LiveCodeBench 21.2%

SciCode 26.5%

Math 500 70.7%

AA Indexes

Intelligence

14.0%

Coding

13.9%

Math

3.7%

Pixtral Large

Mistral

Input: $2.000

Output: $6.000

Pricing

Input (1M): $2.000
Output (1M): $6.000
Blended (3:1): $3.000

Performance

Tokens / s: 30.6
TTFT (token): 0.62s
TTFT (answer): 0.62s

Benchmarks

MMLU Pro 70.1%

GPQA 50.5%

HLE 3.6%

AIME 7.0%

LiveCodeBench 26.1%

SciCode 29.2%

Math 500 71.4%

AA Indexes

Intelligence

14.0%

Coding

—

Math

2.3%

NVIDIA Nemotron Nano 9B V2 (Non-reasoning)

NVIDIA

Input: $0.060

Output: $0.230

Pricing

Input (1M): $0.060
Output (1M): $0.230
Blended (3:1): $0.102

Performance

Tokens / s: 111.8
TTFT (token): 0.54s
TTFT (answer): 0.54s

Benchmarks

MMLU Pro 73.9%

GPQA 55.7%

HLE 4.0%

AIME —

LiveCodeBench 70.1%

SciCode 20.9%

Math 500 —

AA Indexes

Intelligence

13.8%

Coding

7.5%

Math

62.3%

Command A

Cohere

Input: $2.500

Output: $10.000

Pricing

Input (1M): $2.500
Output (1M): $10.000
Blended (3:1): $4.375

Performance

Tokens / s: 52.8
TTFT (token): 0.29s
TTFT (answer): 0.29s

Benchmarks

MMLU Pro 71.2%

GPQA 52.7%

HLE 4.6%

AIME 9.7%

LiveCodeBench 28.7%

SciCode 28.1%

Math 500 81.9%

AA Indexes

Intelligence

13.7%

Coding

9.9%

Math

13.0%

GPT-4 Turbo

OpenAI

Input: $10.000

Output: $30.000

Pricing

Input (1M): $10.000
Output (1M): $30.000
Blended (3:1): $15.000

Performance

Tokens / s: 27.0
TTFT (token): 1.07s
TTFT (answer): 1.07s

Benchmarks

MMLU Pro 69.4%

GPQA —

HLE 3.3%

AIME 15.0%

LiveCodeBench 29.1%

SciCode 31.9%

Math 500 73.7%

AA Indexes

Intelligence

13.7%

Coding

21.5%

Math

—

Hermes 4 - Llama-3.1 70B (Non-reasoning)

Nous Research

Input: $0.130

Output: $0.400

Pricing

Input (1M): $0.130
Output (1M): $0.400
Blended (3:1): $0.198

Performance

Tokens / s: 73.0
TTFT (token): 0.56s
TTFT (answer): 0.56s

Benchmarks

MMLU Pro 66.4%

GPQA 49.1%

HLE 3.6%

AIME —

LiveCodeBench 26.9%

SciCode 27.7%

Math 500 —

AA Indexes

Intelligence

13.6%

Coding

9.2%

Math

11.3%

Aya Expanse 32B

Cohere

Input: $0.500

Output: $1.500

Pricing

Input (1M): $0.500
Output (1M): $1.500
Blended (3:1): $0.750

Performance

Tokens / s: 41.9
TTFT (token): 0.30s
TTFT (answer): 0.30s

Benchmarks

MMLU Pro 37.7%

GPQA 23.0%

HLE 4.5%

AIME —

LiveCodeBench 13.7%

SciCode 14.9%

Math 500 44.9%

AA Indexes

Intelligence

13.6%

Coding

9.8%

Math

2.3%

Nova Pro

Amazon

Input: $0.800

Output: $3.200

Pricing

Input (1M): $0.800
Output (1M): $3.200
Blended (3:1): $1.400

Performance

Tokens / s: —
TTFT (token): —
TTFT (answer): —

Benchmarks

MMLU Pro 69.1%

GPQA 49.9%

HLE 3.4%

AIME 10.7%

LiveCodeBench 23.3%

SciCode 20.8%

Math 500 78.6%

AA Indexes

Intelligence

13.5%

Coding

11.0%

Math

7.0%

Qwen3 14B (Non-reasoning)

Alibaba

Input: $0.350

Output: $1.400

Pricing

Input (1M): $0.350
Output (1M): $1.400
Blended (3:1): $0.613

Performance

Tokens / s: 54.8
TTFT (token): 1.15s
TTFT (answer): 1.15s

Benchmarks

MMLU Pro 67.5%

GPQA 47.0%

HLE 4.2%

AIME 28.0%

LiveCodeBench 28.0%

SciCode 26.5%

Math 500 87.1%

AA Indexes

Intelligence

13.4%

Coding

12.4%

Math

58.0%

GPT-4.1 nano

OpenAI

Input: $0.100

Output: $0.400

Pricing

Input (1M): $0.100
Output (1M): $0.400
Blended (3:1): $0.175

Performance

Tokens / s: 130.1
TTFT (token): 0.38s
TTFT (answer): 0.38s

Benchmarks

MMLU Pro 65.7%

GPQA 51.2%

HLE 3.9%

AIME 23.7%

LiveCodeBench 32.6%

SciCode 25.9%

Math 500 84.8%

AA Indexes

Intelligence

13.4%

Coding

11.2%

Math

24.0%

Phi-4

Microsoft Azure

Input: $0.125

Output: $0.500

Pricing

Input (1M): $0.125
Output (1M): $0.500
Blended (3:1): $0.219

Performance

Tokens / s: 10.9
TTFT (token): 0.58s
TTFT (answer): 0.58s

Benchmarks

MMLU Pro 71.4%

GPQA 57.5%

HLE 4.1%

AIME 14.3%

LiveCodeBench 23.1%

SciCode 26.0%

Math 500 81.0%

AA Indexes

Intelligence

13.2%

Coding

11.2%

Math

18.0%

GLM-4.5V (Non-reasoning)

Z AI

Input: $0.600

Output: $1.800

Pricing

Input (1M): $0.600
Output (1M): $1.800
Blended (3:1): $0.900

Performance

Tokens / s: 28.3
TTFT (token): 0.76s
TTFT (answer): 0.76s

Benchmarks

MMLU Pro 75.1%

GPQA 57.3%

HLE 3.6%

AIME —

LiveCodeBench 35.2%

SciCode 18.8%

Math 500 —

AA Indexes

Intelligence

13.2%

Coding

10.8%

Math

15.3%

Gemini 2.5 Flash-Lite (Non-reasoning)

Google

Input: $0.100

Output: $0.400

Pricing

Input (1M): $0.100
Output (1M): $0.400
Blended (3:1): $0.175

Performance

Tokens / s: 256.1
TTFT (token): 0.36s
TTFT (answer): 0.36s

Benchmarks

MMLU Pro 72.4%

GPQA 47.4%

HLE 3.7%

AIME 50.0%

LiveCodeBench 40.0%

SciCode 17.7%

Math 500 92.6%

AA Indexes

Intelligence

13.1%

Coding

7.4%

Math

35.3%

Llama 3.1 Instruct 70B

Meta

Input: $0.560

Output: $0.560

Pricing

Input (1M): $0.560
Output (1M): $0.560
Blended (3:1): $0.560

Performance

Tokens / s: 61.5
TTFT (token): 0.40s
TTFT (answer): 0.40s

Benchmarks

MMLU Pro 67.6%

GPQA 40.9%

HLE 4.6%

AIME 17.3%

LiveCodeBench 23.2%

SciCode 26.7%

Math 500 64.9%

AA Indexes

Intelligence

13.1%

Coding

10.9%

Math

4.0%

Qwen3 1.7B (Reasoning)

Alibaba

Input: $0.110

Output: $1.260

Pricing

Input (1M): $0.110
Output (1M): $1.260
Blended (3:1): $0.398

Performance

Tokens / s: 124.8
TTFT (token): 0.93s
TTFT (answer): 16.96s

Benchmarks

MMLU Pro 57.0%

GPQA 35.6%

HLE 4.8%

AIME 51.0%

LiveCodeBench 30.8%

SciCode 4.3%

Math 500 89.4%

AA Indexes

Intelligence

13.1%

Coding

1.4%

Math

38.7%

Mistral Large 2 (Jul '24)

Mistral

Input: $2.000

Output: $6.000

Pricing

Input (1M): $2.000
Output (1M): $6.000
Blended (3:1): $3.000

Performance

Tokens / s: —
TTFT (token): —
TTFT (answer): —

Benchmarks

MMLU Pro 68.3%

GPQA 47.2%

HLE 3.2%

AIME 9.3%

LiveCodeBench 26.7%

SciCode 27.1%

Math 500 71.4%

AA Indexes

Intelligence

13.0%

Coding

—

Math

—

Qwen2.5 Coder Instruct 32B

Alibaba

Input: $0.130

Output: $0.175

Pricing

Input (1M): $0.130
Output (1M): $0.175
Blended (3:1): $0.141

Performance

Tokens / s: 35.7
TTFT (token): 0.51s
TTFT (answer): 0.51s

Benchmarks

MMLU Pro 63.5%

GPQA 41.7%

HLE 3.8%

AIME 12.0%

LiveCodeBench 29.5%

SciCode 27.1%

Math 500 76.7%

AA Indexes

Intelligence

12.9%

Coding

—

Math

—

GPT-4

OpenAI

Input: $30.000

Output: $60.000

Pricing

Input (1M): $30.000
Output (1M): $60.000
Blended (3:1): $37.500

Performance

Tokens / s: 28.3
TTFT (token): 0.68s
TTFT (answer): 0.68s

Benchmarks

MMLU Pro —

GPQA —

HLE —

AIME —

LiveCodeBench —

SciCode —

Math 500 —

AA Indexes

Intelligence

12.8%

Coding

13.1%

Math

—

Mistral Small 3

Mistral

Input: $0.100

Output: $0.300

Pricing

Input (1M): $0.100
Output (1M): $0.300
Blended (3:1): $0.150

Performance

Tokens / s: 231.4
TTFT (token): 0.34s
TTFT (answer): 0.34s

Benchmarks

MMLU Pro 65.2%

GPQA 46.2%

HLE 4.1%

AIME 8.0%

LiveCodeBench 25.2%

SciCode 23.6%

Math 500 71.5%

AA Indexes

Intelligence

12.7%

Coding

—

Math

4.3%

GPT-4o mini

OpenAI

Input: $0.150

Output: $0.600

Pricing

Input (1M): $0.150
Output (1M): $0.600
Blended (3:1): $0.263

Performance

Tokens / s: 49.0
TTFT (token): 0.53s
TTFT (answer): 0.53s

Benchmarks

MMLU Pro 64.8%

GPQA 42.6%

HLE 4.0%

AIME 11.7%

LiveCodeBench 23.4%

SciCode 22.9%

Math 500 78.9%

AA Indexes

Intelligence

12.6%

Coding

—

Math

14.7%

Qwen3 4B (Non-reasoning)

Alibaba

Input: $0.110

Output: $0.420

Pricing

Input (1M): $0.110
Output (1M): $0.420
Blended (3:1): $0.188

Performance

Tokens / s: 85.1
TTFT (token): 0.94s
TTFT (answer): 0.94s

Benchmarks

MMLU Pro 58.6%

GPQA 39.8%

HLE 3.7%

AIME 21.3%

LiveCodeBench 23.3%

SciCode 16.7%

Math 500 84.3%

AA Indexes

Intelligence

12.5%

Coding

—

Math

—

Claude 3 Opus

Anthropic

Input: $15.000

Output: $75.000

Pricing

Input (1M): $15.000
Output (1M): $75.000
Blended (3:1): $30.000

Performance

Tokens / s: —
TTFT (token): —
TTFT (answer): —

Benchmarks

MMLU Pro 69.6%

GPQA 48.9%

HLE 3.1%

AIME 3.3%

LiveCodeBench 27.9%

SciCode 23.3%

Math 500 64.1%

AA Indexes

Intelligence

12.5%

Coding

19.5%

Math

—

Nova Lite

Amazon

Input: $0.060

Output: $0.240

Pricing

Input (1M): $0.060
Output (1M): $0.240
Blended (3:1): $0.105

Performance

Tokens / s: 238.3
TTFT (token): 0.38s
TTFT (answer): 0.38s

Benchmarks

MMLU Pro 59.0%

GPQA 43.3%

HLE 4.6%

AIME 10.7%

LiveCodeBench 16.7%

SciCode 13.9%

Math 500 76.5%

AA Indexes

Intelligence

12.5%

Coding

5.1%

Math

7.0%

Qwen3 30B A3B (Non-reasoning)

Alibaba

Input: $0.200

Output: $0.800

Pricing

Input (1M): $0.200
Output (1M): $0.800
Blended (3:1): $0.350

Performance

Tokens / s: 61.7
TTFT (token): 1.07s
TTFT (answer): 1.07s

Benchmarks

MMLU Pro 71.0%

GPQA 51.5%

HLE 4.6%

AIME 26.0%

LiveCodeBench 32.2%

SciCode 26.4%

Math 500 86.3%

AA Indexes

Intelligence

12.4%

Coding

13.3%

Math

21.7%

Ministral 8B

Mistral

Input: $0.100

Output: $0.100

Pricing

Input (1M): $0.100
Output (1M): $0.100
Blended (3:1): $0.100

Performance

Tokens / s: 195.9
TTFT (token): 0.34s
TTFT (answer): 0.34s

Benchmarks

MMLU Pro 38.9%

GPQA 27.6%

HLE 4.9%

AIME 3.7%

LiveCodeBench 11.2%

SciCode 11.5%

Math 500 57.1%

AA Indexes

Intelligence

12.4%

Coding

7.6%

Math

3.0%

Llama 3.1 Instruct 8B

Meta

Input: $0.100

Output: $0.100

Pricing

Input (1M): $0.100
Output (1M): $0.100
Blended (3:1): $0.100

Performance

Tokens / s: 162.8
TTFT (token): 0.36s
TTFT (answer): 0.36s

Benchmarks

MMLU Pro 47.6%

GPQA 25.9%

HLE 5.1%

AIME 7.7%

LiveCodeBench 11.6%

SciCode 13.2%

Math 500 51.9%

AA Indexes

Intelligence

12.2%

Coding

4.9%

Math

4.3%

Ministral 3 3B

Mistral

Input: $0.100

Output: $0.100

Pricing

Input (1M): $0.100
Output (1M): $0.100
Blended (3:1): $0.100

Performance

Tokens / s: 293.0
TTFT (token): 0.27s
TTFT (answer): 0.27s

Benchmarks

MMLU Pro 52.4%

GPQA 35.8%

HLE 5.3%

AIME —

LiveCodeBench 24.7%

SciCode 14.4%

Math 500 —

AA Indexes

Intelligence

12.1%

Coding

4.8%

Math

22.0%

Olmo 3.1 32B Instruct

Allen Institute for AI

Input: $0.200

Output: $0.600

Pricing

Input (1M): $0.200
Output (1M): $0.600
Blended (3:1): $0.300

Performance

Tokens / s: 47.6
TTFT (token): 0.26s
TTFT (answer): 0.26s

Benchmarks

MMLU Pro —

GPQA 53.9%

HLE 4.9%

AIME —

LiveCodeBench —

SciCode 16.7%

Math 500 —

AA Indexes

Intelligence

12.0%

Coding

5.6%

Math

—

Reka Flash (Sep '24)

Reka AI

Input: $0.200

Output: $0.800

Pricing

Input (1M): $0.200
Output (1M): $0.800
Blended (3:1): $0.350

Performance

Tokens / s: 71.1
TTFT (token): 1.28s
TTFT (answer): 1.28s

Benchmarks

MMLU Pro —

GPQA —

HLE —

AIME —

LiveCodeBench —

SciCode —

Math 500 52.9%

AA Indexes

Intelligence

12.0%

Coding

—

Math

—

EXAONE 4.0 32B (Non-reasoning)

LG AI Research

Input: $0.600

Output: $1.000

Pricing

Input (1M): $0.600
Output (1M): $1.000
Blended (3:1): $0.700

Performance

Tokens / s: 88.3
TTFT (token): 0.31s
TTFT (answer): 0.31s

Benchmarks

MMLU Pro 76.8%

GPQA 62.8%

HLE 4.9%

AIME 47.0%

LiveCodeBench 47.2%

SciCode 25.2%

Math 500 93.9%

AA Indexes

Intelligence

12.0%

Coding

9.4%

Math

39.3%

Qwen2.5 Turbo

Alibaba

Input: $0.050

Output: $0.200

Pricing

Input (1M): $0.050
Output (1M): $0.200
Blended (3:1): $0.087

Performance

Tokens / s: 66.4
TTFT (token): 1.02s
TTFT (answer): 1.02s

Benchmarks

MMLU Pro 63.3%

GPQA 41.0%

HLE 4.2%

AIME 12.0%

LiveCodeBench 16.3%

SciCode 15.3%

Math 500 80.5%

AA Indexes

Intelligence

12.0%

Coding

—

Math

—

Llama 3.2 Instruct 90B (Vision)

Meta

Input: $0.720

Output: $0.720

Pricing

Input (1M): $0.720
Output (1M): $0.720
Blended (3:1): $0.720

Performance

Tokens / s: 36.6
TTFT (token): 0.35s
TTFT (answer): 0.35s

Benchmarks

MMLU Pro 67.1%

GPQA 43.2%

HLE 4.9%

AIME 5.0%

LiveCodeBench 21.4%

SciCode 24.0%

Math 500 62.9%

AA Indexes

Intelligence

11.9%

Coding

—

Math

—

Solar Mini

Upstage

Input: $0.150

Output: $0.150

Pricing

Input (1M): $0.150
Output (1M): $0.150
Blended (3:1): $0.150

Performance

Tokens / s: 84.5
TTFT (token): 1.01s
TTFT (answer): 1.01s

Benchmarks

MMLU Pro —

GPQA —

HLE —

AIME —

LiveCodeBench —

SciCode —

Math 500 33.1%

AA Indexes

Intelligence

11.9%

Coding

—

Math

—

Granite 4.0 H Small

IBM

Input: $0.060

Output: $0.250

Pricing

Input (1M): $0.060
Output (1M): $0.250
Blended (3:1): $0.107

Performance

Tokens / s: 370.8
TTFT (token): 8.80s
TTFT (answer): 8.80s

Benchmarks

MMLU Pro 62.4%

GPQA 41.6%

HLE 3.7%

AIME —

LiveCodeBench 25.1%

SciCode 20.9%

Math 500 —

AA Indexes

Intelligence

11.5%

Coding

8.5%

Math

13.7%

CompactifAI Llama 3.1 8B Slim

Multiverse Computing

Input: $0.050

Output: $0.070

Pricing

Input (1M): $0.050
Output (1M): $0.070
Blended (3:1): $0.055

Performance

Tokens / s: 228.4
TTFT (token): 0.24s
TTFT (answer): 0.24s

Benchmarks

MMLU Pro 32.1%

GPQA 22.1%

HLE 5.5%

AIME —

LiveCodeBench 11.3%

SciCode 6.2%

Math 500 —

AA Indexes

Intelligence

11.2%

Coding

5.9%

Math

6.7%

Llama 3.2 Instruct 11B (Vision)

Meta

Input: $0.160

Output: $0.160

Pricing

Input (1M): $0.160
Output (1M): $0.160
Blended (3:1): $0.160

Performance

Tokens / s: 48.8
TTFT (token): 0.41s
TTFT (answer): 0.41s

Benchmarks

MMLU Pro 46.4%

GPQA 22.1%

HLE 5.2%

AIME 9.3%

LiveCodeBench 11.0%

SciCode 11.2%

Math 500 51.6%

AA Indexes

Intelligence

10.9%

Coding

4.3%

Math

1.7%

Ministral 3B

Mistral

Input: $0.040

Output: $0.040

Pricing

Input (1M): $0.040
Output (1M): $0.040
Blended (3:1): $0.040

Performance

Tokens / s: 272.6
TTFT (token): 0.33s
TTFT (answer): 0.33s

Benchmarks

MMLU Pro 33.9%

GPQA 26.0%

HLE 5.5%

AIME —

LiveCodeBench 6.9%

SciCode 9.4%

Math 500 53.7%

AA Indexes

Intelligence

10.9%

Coding

5.4%

Math

30.0%

Granite 3.3 8B (Non-reasoning)

IBM

Input: $0.030

Output: $0.250

Pricing

Input (1M): $0.030
Output (1M): $0.250
Blended (3:1): $0.085

Performance

Tokens / s: 488.2
TTFT (token): 7.57s
TTFT (answer): 7.57s

Benchmarks

MMLU Pro 46.8%

GPQA 33.8%

HLE 4.2%

AIME 4.7%

LiveCodeBench 12.7%

SciCode 10.1%

Math 500 66.5%

AA Indexes

Intelligence

10.8%

Coding

3.4%

Math

6.7%

Qwen3 8B (Non-reasoning)

Alibaba

Input: $0.180

Output: $0.700

Pricing

Input (1M): $0.180
Output (1M): $0.700
Blended (3:1): $0.310

Performance

Tokens / s: 79.7
TTFT (token): 0.92s
TTFT (answer): 0.92s

Benchmarks

MMLU Pro 64.3%

GPQA 45.2%

HLE 2.8%

AIME 24.3%

LiveCodeBench 20.2%

SciCode 16.8%

Math 500 82.8%

AA Indexes

Intelligence

10.8%

Coding

7.1%

Math

24.3%

Qwen3 Omni 30B A3B Instruct

Alibaba

Input: $0.250

Output: $0.970

Pricing

Input (1M): $0.250
Output (1M): $0.970
Blended (3:1): $0.430

Performance

Tokens / s: 88.3
TTFT (token): 0.91s
TTFT (answer): 0.91s

Benchmarks

MMLU Pro 72.5%

GPQA 62.0%

HLE 5.1%

AIME —

LiveCodeBench 42.2%

SciCode 18.6%

Math 500 —

AA Indexes

Intelligence

10.7%

Coding

7.2%

Math

52.3%

Jamba 1.5 Large

AI21 Labs

Input: $2.000

Output: $8.000

Pricing

Input (1M): $2.000
Output (1M): $8.000
Blended (3:1): $3.500

Performance

Tokens / s: —
TTFT (token): —
TTFT (answer): —

Benchmarks

MMLU Pro 57.2%

GPQA 42.7%

HLE 4.0%

AIME 4.7%

LiveCodeBench 14.3%

SciCode 16.3%

Math 500 60.6%

AA Indexes

Intelligence

10.7%

Coding

—

Math

—

Jamba 1.6 Large

AI21 Labs

Input: $2.000

Output: $8.000

Pricing

Input (1M): $2.000
Output (1M): $8.000
Blended (3:1): $3.500

Performance

Tokens / s: 44.9
TTFT (token): 0.82s
TTFT (answer): 0.82s

Benchmarks

MMLU Pro 56.5%

GPQA 38.7%

HLE 4.0%

AIME 4.7%

LiveCodeBench 17.2%

SciCode 18.4%

Math 500 58.0%

AA Indexes

Intelligence

10.6%

Coding

—

Math

—

Hermes 3 - Llama-3.1 70B

Nous Research

Input: $0.300

Output: $0.300

Pricing

Input (1M): $0.300
Output (1M): $0.300
Blended (3:1): $0.300

Performance

Tokens / s: 41.3
TTFT (token): 0.29s
TTFT (answer): 0.29s

Benchmarks

MMLU Pro 57.1%

GPQA 40.1%

HLE 4.1%

AIME 2.3%

LiveCodeBench 18.8%

SciCode 23.1%

Math 500 53.8%

AA Indexes

Intelligence

10.6%

Coding

—

Math

—

NVIDIA Nemotron Nano 12B v2 VL (Non-reasoning)

NVIDIA

Input: $0.200

Output: $0.600

Pricing

Input (1M): $0.200
Output (1M): $0.600
Blended (3:1): $0.300

Performance

Tokens / s: 131.2
TTFT (token): 0.56s
TTFT (answer): 0.56s

Benchmarks

MMLU Pro 64.9%

GPQA 43.9%

HLE 4.5%

AIME —

LiveCodeBench 34.5%

SciCode 17.6%

Math 500 —

AA Indexes

Intelligence

10.4%

Coding

5.9%

Math

26.7%

Nova Micro

Amazon

Input: $0.035

Output: $0.140

Pricing

Input (1M): $0.035
Output (1M): $0.140
Blended (3:1): $0.061

Performance

Tokens / s: 421.4
TTFT (token): 0.36s
TTFT (answer): 0.36s

Benchmarks

MMLU Pro 53.1%

GPQA 35.8%

HLE 4.7%

AIME 8.0%

LiveCodeBench 14.0%

SciCode 9.4%

Math 500 70.3%

AA Indexes

Intelligence

10.3%

Coding

4.1%

Math

6.0%

Claude 3 Sonnet

Anthropic

Input: $3.000

Output: $15.000

Pricing

Input (1M): $3.000
Output (1M): $15.000
Blended (3:1): $6.000

Performance

Tokens / s: —
TTFT (token): —
TTFT (answer): —

Benchmarks

MMLU Pro 57.9%

GPQA 40.0%

HLE 3.8%

AIME 4.7%

LiveCodeBench 17.5%

SciCode 22.9%

Math 500 41.4%

AA Indexes

Intelligence

10.3%

Coding

—

Math

—

Mistral Small (Sep '24)

Mistral

Input: $0.200

Output: $0.600

Pricing

Input (1M): $0.200
Output (1M): $0.600
Blended (3:1): $0.300

Performance

Tokens / s: 110.9
TTFT (token): 0.30s
TTFT (answer): 0.30s

Benchmarks

MMLU Pro 52.9%

GPQA 38.1%

HLE 4.3%

AIME 6.3%

LiveCodeBench 14.1%

SciCode 15.6%

Math 500 56.3%

AA Indexes

Intelligence

10.2%

Coding

—

Math

—

Llama 3 Instruct 70B

Meta

Input: $0.510

Output: $0.740

Pricing

Input (1M): $0.510
Output (1M): $0.740
Blended (3:1): $0.568

Performance

Tokens / s: 38.0
TTFT (token): 0.43s
TTFT (answer): 0.43s

Benchmarks

MMLU Pro 57.4%

GPQA 37.9%

HLE 4.4%

AIME —

LiveCodeBench 19.8%

SciCode 18.9%

Math 500 48.3%

AA Indexes

Intelligence

10.2%

Coding

6.8%

Math

—

Phi-3 Mini Instruct 3.8B

Microsoft Azure

Input: $0.130

Output: $0.520

Pricing

Input (1M): $0.130
Output (1M): $0.520
Blended (3:1): $0.228

Performance

Tokens / s: —
TTFT (token): —
TTFT (answer): —

Benchmarks

MMLU Pro 43.5%

GPQA 31.9%

HLE 4.4%

AIME 4.0%

LiveCodeBench 11.6%

SciCode 9.0%

Math 500 45.7%

AA Indexes

Intelligence

10.1%

Coding

3.0%

Math

30.0%

Aya Expanse 8B

Cohere

Input: $0.500

Output: $1.500

Pricing

Input (1M): $0.500
Output (1M): $1.500
Blended (3:1): $0.750

Performance

Tokens / s: 79.5
TTFT (token): 0.26s
TTFT (answer): 0.26s

Benchmarks

MMLU Pro 31.2%

GPQA 24.7%

HLE 5.1%

AIME —

LiveCodeBench 7.0%

SciCode 7.8%

Math 500 32.1%

AA Indexes

Intelligence

10.0%

Coding

4.9%

Math

—

Mistral Large (Feb '24)

Mistral

Input: $4.000

Output: $12.000

Pricing

Input (1M): $4.000
Output (1M): $12.000
Blended (3:1): $6.000

Performance

Tokens / s: —
TTFT (token): —
TTFT (answer): —

Benchmarks

MMLU Pro 51.5%

GPQA 35.1%

HLE 3.4%

AIME —

LiveCodeBench 17.8%

SciCode 20.8%

Math 500 52.7%

AA Indexes

Intelligence

9.9%

Coding

—

Math

—

Llama 3.2 Instruct 3B

Meta

Input: $0.060

Output: $0.060

Pricing

Input (1M): $0.060
Output (1M): $0.060
Blended (3:1): $0.060

Performance

Tokens / s: 46.5
TTFT (token): 0.50s
TTFT (answer): 0.50s

Benchmarks

MMLU Pro 34.7%

GPQA 25.5%

HLE 5.2%

AIME 6.7%

LiveCodeBench 8.3%

SciCode 5.2%

Math 500 48.9%

AA Indexes

Intelligence

9.7%

Coding

—

Math

3.3%

Llama 2 Chat 7B

Meta

Input: $0.050

Output: $0.250

Pricing

Input (1M): $0.050
Output (1M): $0.250
Blended (3:1): $0.100

Performance

Tokens / s: 109.5
TTFT (token): 0.54s
TTFT (answer): 0.54s

Benchmarks

MMLU Pro 16.4%

GPQA 22.7%

HLE 5.8%

AIME —

LiveCodeBench 0.2%

SciCode —

Math 500 5.9%

AA Indexes

Intelligence

9.7%

Coding

—

Math

—

Jamba 1.7 Large

AI21 Labs

Input: $2.000

Output: $8.000

Pricing

Input (1M): $2.000
Output (1M): $8.000
Blended (3:1): $3.500

Performance

Tokens / s: 41.9
TTFT (token): 0.77s
TTFT (answer): 0.77s

Benchmarks

MMLU Pro 57.7%

GPQA 39.0%

HLE 3.8%

AIME 5.7%

LiveCodeBench 18.1%

SciCode 18.8%

Math 500 60.0%

AA Indexes

Intelligence

9.7%

Coding

7.8%

Math

2.3%

Claude 3 Haiku

Anthropic

Input: $0.250

Output: $1.250

Pricing

Input (1M): $0.250
Output (1M): $1.250
Blended (3:1): $0.500

Performance

Tokens / s: 111.8
TTFT (token): 0.34s
TTFT (answer): 0.34s

Benchmarks

MMLU Pro —

GPQA —

HLE —

AIME 1.0%

LiveCodeBench 15.4%

SciCode 18.6%

Math 500 39.4%

AA Indexes

Intelligence

9.3%

Coding

—

Math

—

Llama 3.2 Instruct 1B

Meta

Input: $0.053

Output: $0.055

Pricing

Input (1M): $0.053
Output (1M): $0.055
Blended (3:1): $0.053

Performance

Tokens / s: 74.5
TTFT (token): 0.55s
TTFT (answer): 0.55s

Benchmarks

MMLU Pro 20.0%

GPQA 19.6%

HLE 5.3%

AIME —

LiveCodeBench 1.9%

SciCode 1.7%

Math 500 14.0%

AA Indexes

Intelligence

9.1%

Coding

60.0%

Math

—

Mistral Small (Feb '24)

Mistral

Input: $1.000

Output: $3.000

Pricing

Input (1M): $1.000
Output (1M): $3.000
Blended (3:1): $1.500

Performance

Tokens / s: 108.1
TTFT (token): 0.29s
TTFT (answer): 0.29s

Benchmarks

MMLU Pro 41.9%

GPQA 30.2%

HLE 4.4%

AIME 0.7%

LiveCodeBench 11.1%

SciCode 13.4%

Math 500 56.2%

AA Indexes

Intelligence

9.0%

Coding

—

Math

—

GPT-3.5 Turbo

OpenAI

Input: $0.500

Output: $1.500

Pricing

Input (1M): $0.500
Output (1M): $1.500
Blended (3:1): $0.750

Performance

Tokens / s: 112.3
TTFT (token): 0.43s
TTFT (answer): 0.43s

Benchmarks

MMLU Pro 46.2%

GPQA 29.7%

HLE —

AIME —

LiveCodeBench —

SciCode —

Math 500 44.1%

AA Indexes

Intelligence

9.0%

Coding

10.7%

Math

—

Mistral Medium

Mistral

Input: $2.750

Output: $8.100

Pricing

Input (1M): $2.750
Output (1M): $8.100
Blended (3:1): $4.088

Performance

Tokens / s: 92.0
TTFT (token): 0.40s
TTFT (answer): 0.40s

Benchmarks

MMLU Pro 49.1%

GPQA 34.9%

HLE 3.4%

AIME 3.7%

LiveCodeBench 9.9%

SciCode 11.8%

Math 500 40.5%

AA Indexes

Intelligence

9.0%

Coding

—

Math

—

Pixtral 12B (2409)

Mistral

Input: $0.150

Output: $0.150

Pricing

Input (1M): $0.150
Output (1M): $0.150
Blended (3:1): $0.150

Performance

Tokens / s: 136.6
TTFT (token): 0.43s
TTFT (answer): 0.43s

Benchmarks

MMLU Pro 47.3%

GPQA 34.3%

HLE 5.3%

AIME —

LiveCodeBench 11.5%

SciCode 13.5%

Math 500 45.8%

AA Indexes

Intelligence

8.9%

Coding

—

Math

—

Llama 3 Instruct 8B

Meta

Input: $0.045

Output: $0.155

Pricing

Input (1M): $0.045
Output (1M): $0.155
Blended (3:1): $0.070

Performance

Tokens / s: 67.1
TTFT (token): 0.37s
TTFT (answer): 0.37s

Benchmarks

MMLU Pro 40.5%

GPQA 29.6%

HLE 5.1%

AIME —

LiveCodeBench 9.6%

SciCode 11.9%

Math 500 49.9%

AA Indexes

Intelligence

8.7%

Coding

4.0%

Math

—

Command-R+ (Apr '24)

Cohere

Input: $3.000

Output: $15.000

Pricing

Input (1M): $3.000
Output (1M): $15.000
Blended (3:1): $6.000

Performance

Tokens / s: —
TTFT (token): —
TTFT (answer): —

Benchmarks

MMLU Pro 43.2%

GPQA 32.3%

HLE 4.5%

AIME 0.7%

LiveCodeBench 12.2%

SciCode 11.8%

Math 500 27.9%

AA Indexes

Intelligence

8.3%

Coding

—

Math

—

Olmo 3 7B Instruct

Allen Institute for AI

Input: $0.100

Output: $0.200

Pricing

Input (1M): $0.100
Output (1M): $0.200
Blended (3:1): $0.125

Performance

Tokens / s: 37.0
TTFT (token): 0.44s
TTFT (answer): 0.44s

Benchmarks

MMLU Pro 52.2%

GPQA 40.0%

HLE 5.8%

AIME —

LiveCodeBench 26.6%

SciCode 10.3%

Math 500 —

AA Indexes

Intelligence

8.2%

Coding

3.4%

Math

41.3%

Jamba 1.5 Mini

AI21 Labs

Input: $0.200

Output: $0.400

Pricing

Input (1M): $0.200
Output (1M): $0.400
Blended (3:1): $0.250

Performance

Tokens / s: —
TTFT (token): —
TTFT (answer): —

Benchmarks

MMLU Pro 37.1%

GPQA 30.2%

HLE 5.1%

AIME 1.0%

LiveCodeBench 6.2%

SciCode 8.0%

Math 500 35.7%

AA Indexes

Intelligence

8.0%

Coding

—

Math

—

Jamba 1.6 Mini

AI21 Labs

Input: $0.200

Output: $0.400

Pricing

Input (1M): $0.200
Output (1M): $0.400
Blended (3:1): $0.250

Performance

Tokens / s: 127.1
TTFT (token): 0.63s
TTFT (answer): 0.63s

Benchmarks

MMLU Pro 36.7%

GPQA 30.0%

HLE 4.6%

AIME 3.3%

LiveCodeBench 7.1%

SciCode 10.1%

Math 500 25.7%

AA Indexes

Intelligence

7.9%

Coding

—

Math

—

Jamba 1.7 Mini

AI21 Labs

Input: $0.200

Output: $0.400

Pricing

Input (1M): $0.200
Output (1M): $0.400
Blended (3:1): $0.250

Performance

Tokens / s: 158.1
TTFT (token): 0.65s
TTFT (answer): 0.65s

Benchmarks

MMLU Pro 38.8%

GPQA 32.2%

HLE 4.5%

AIME 1.3%

LiveCodeBench 6.1%

SciCode 9.3%

Math 500 25.8%

AA Indexes

Intelligence

7.9%

Coding

3.1%

Math

30.0%

Gemma 2 9B

Google

Input: $0.030

Output: $0.090

Pricing

Input (1M): $0.030
Output (1M): $0.090
Blended (3:1): $0.045

Performance

Tokens / s: —
TTFT (token): —
TTFT (answer): —

Benchmarks

MMLU Pro 49.5%

GPQA 31.1%

HLE 3.9%

AIME —

LiveCodeBench 12.6%

SciCode 0.7%

Math 500 51.7%

AA Indexes

Intelligence

7.8%

Coding

—

Math

—

Mixtral 8x7B Instruct

Mistral

Input: $0.540

Output: $0.600

Pricing

Input (1M): $0.540
Output (1M): $0.600
Blended (3:1): $0.540

Performance

Tokens / s: —
TTFT (token): —
TTFT (answer): —

Benchmarks

MMLU Pro 38.7%

GPQA 29.2%

HLE 4.5%

AIME —

LiveCodeBench 6.6%

SciCode 2.8%

Math 500 29.9%

AA Indexes

Intelligence

7.7%

Coding

—

Math

—

Command-R (Mar '24)

Cohere

Input: $0.500

Output: $1.500

Pricing

Input (1M): $0.500
Output (1M): $1.500
Blended (3:1): $0.750

Performance

Tokens / s: —
TTFT (token): —
TTFT (answer): —

Benchmarks

MMLU Pro 33.8%

GPQA 28.4%

HLE 4.8%

AIME 0.7%

LiveCodeBench 4.8%

SciCode 6.2%

Math 500 16.4%

AA Indexes

Intelligence

7.4%

Coding

—

Math

—

Mistral 7B Instruct

Mistral

Input: $0.250

Output: $0.250

Pricing

Input (1M): $0.250
Output (1M): $0.250
Blended (3:1): $0.250

Performance

Tokens / s: 125.6
TTFT (token): 0.34s
TTFT (answer): 0.34s

Benchmarks

MMLU Pro 24.5%

GPQA 17.7%

HLE 4.3%

AIME —

LiveCodeBench 4.6%

SciCode 2.4%

Math 500 12.1%

AA Indexes

Intelligence

7.4%

Coding

—

Math

—

Command-R+ (Aug '24)

Cohere

Input: $2.500

Output: $10.000

Pricing

Input (1M): $2.500
Output (1M): $10.000
Blended (3:1): $4.375

Performance

Tokens / s: 21.0
TTFT (token): 0.56s
TTFT (answer): 0.56s

Benchmarks

MMLU Pro 42.7%

GPQA 33.7%

HLE 5.0%

AIME —

LiveCodeBench 11.1%

SciCode 12.2%

Math 500 40.2%

AA Indexes

Intelligence

7.1%

Coding

—

Math

—

Qwen3 1.7B (Non-reasoning)

Alibaba

Input: $0.110

Output: $0.420

Pricing

Input (1M): $0.110
Output (1M): $0.420
Blended (3:1): $0.188

Performance

Tokens / s: 116.5
TTFT (token): 0.91s
TTFT (answer): 0.91s

Benchmarks

MMLU Pro 41.1%

GPQA 28.3%

HLE 5.2%

AIME 9.7%

LiveCodeBench 12.6%

SciCode 6.9%

Math 500 71.7%

AA Indexes

Intelligence

6.8%

Coding

2.3%

Math

7.3%

Qwen3 0.6B (Reasoning)

Alibaba

Input: $0.110

Output: $1.260

Pricing

Input (1M): $0.110
Output (1M): $1.260
Blended (3:1): $0.398

Performance

Tokens / s: 199.0
TTFT (token): 0.90s
TTFT (answer): 10.95s

Benchmarks

MMLU Pro 34.7%

GPQA 23.9%

HLE 5.7%

AIME 10.0%

LiveCodeBench 12.1%

SciCode 2.8%

Math 500 75.0%

AA Indexes

Intelligence

6.4%

Coding

90.0%

Math

18.0%

Gemma 3n E4B Instruct

Google

Input: $0.020

Output: $0.040

Pricing

Input (1M): $0.020
Output (1M): $0.040
Blended (3:1): $0.025

Performance

Tokens / s: 60.5
TTFT (token): 0.33s
TTFT (answer): 0.33s

Benchmarks

MMLU Pro 48.8%

GPQA 29.6%

HLE 4.4%

AIME 13.7%

LiveCodeBench 14.6%

SciCode 8.1%

Math 500 77.1%

AA Indexes

Intelligence

6.3%

Coding

4.2%

Math

14.3%

Qwen3 0.6B (Non-reasoning)

Alibaba

Input: $0.110

Output: $0.420

Pricing

Input (1M): $0.110
Output (1M): $0.420
Blended (3:1): $0.188

Performance

Tokens / s: 190.1
TTFT (token): 0.94s
TTFT (answer): 0.94s

Benchmarks

MMLU Pro 23.1%

GPQA 23.1%

HLE 5.2%

AIME 1.7%

LiveCodeBench 7.3%

SciCode 4.1%

Math 500 52.1%

AA Indexes

Intelligence

6.1%

Coding

1.4%

Math

10.3%

Mistral NeMo

Mistral

Input: $0.150

Output: $0.150

Pricing

Input (1M): $0.150
Output (1M): $0.150
Blended (3:1): $0.150

Performance

Tokens / s: 190.2
TTFT (token): 0.34s
TTFT (answer): 0.34s

Benchmarks

MMLU Pro 39.9%

GPQA 31.4%

HLE 4.4%

AIME 0.3%

LiveCodeBench 5.7%

SciCode 10.4%

Math 500 39.5%

AA Indexes

Intelligence

5.2%

Coding

—

Math

—

Command-R (Aug '24)

Cohere

Input: $0.150

Output: $0.600

Pricing

Input (1M): $0.150
Output (1M): $0.600
Blended (3:1): $0.263

Performance

Tokens / s: 59.8
TTFT (token): 0.24s
TTFT (answer): 0.24s

Benchmarks

MMLU Pro 33.7%

GPQA 28.9%

HLE 5.1%

AIME 0.3%

LiveCodeBench 4.4%

SciCode 8.7%

Math 500 14.9%

AA Indexes

Intelligence

100.0%

Coding

—

Math

—

Cogito v2.1 (Reasoning)

Deep Cogito

Input: $1.250

Output: $1.250

Pricing

Input (1M): $1.250
Output (1M): $1.250
Blended (3:1): $1.250

Performance

Tokens / s: 73.9
TTFT (token): 0.32s
TTFT (answer): 27.39s

Benchmarks

MMLU Pro 84.9%

GPQA 76.8%

HLE 11.0%

AIME —

LiveCodeBench 68.8%

SciCode 41.0%

Math 500 —

AA Indexes

Intelligence

—

Coding

24.8%

Math

72.7%

DeepSeek-OCR

DeepSeek

Input: $0.030

Output: $0.100

Pricing

Input (1M): $0.030
Output (1M): $0.100
Blended (3:1): $0.048

Performance

Tokens / s: 306.8
TTFT (token): 0.37s
TTFT (answer): 0.37s

Benchmarks

MMLU Pro —

GPQA —

HLE —

AIME —

LiveCodeBench —

SciCode —

Math 500 —

AA Indexes

Intelligence

—

Coding

—

Math

—

Grok 3 mini Reasoning (Low)

xAI

Input: $0.300

Output: $0.500

Pricing

Input (1M): $0.300
Output (1M): $0.500
Blended (3:1): $0.350

Performance

Tokens / s: 110.8
TTFT (token): 0.56s
TTFT (answer): 18.62s

Benchmarks

MMLU Pro —

GPQA —

HLE —

AIME —

LiveCodeBench —

SciCode —

Math 500 —

AA Indexes

Intelligence

—

Coding

—

Math

—

Doubao-Seed-1.8

ByteDance Seed

Input: $0.110

Output: $0.280

Pricing

Input (1M): $0.110
Output (1M): $0.280
Blended (3:1): $0.152

Performance

Tokens / s: 43.1
TTFT (token): 2.34s
TTFT (answer): 48.71s

Benchmarks

MMLU Pro —

GPQA —

HLE —

AIME —

LiveCodeBench —

SciCode —

Math 500 —

AA Indexes

Intelligence

—

Coding

—

Math

—

Model	Provider	Input	Output	MMLU	GPQA	TPS
GPT-5.2 (xhigh)	OpenAI	$1.750	$14.000	87.4%	90.3%	101.7
Claude Opus 4.5 (Reasoning)	Anthropic	$5.000	$25.000	89.5%	86.6%	83.5
Gemini 3 Pro Preview (high)	Google	$2.000	$12.000	89.8%	90.8%	116.3
GPT-5.1 (high)	OpenAI	$1.250	$10.000	87.0%	87.3%	93.4
Gemini 3 Flash Preview (Reasoning)	Google	$0.500	$3.000	89.0%	89.8%	200.6
GPT-5.2 (medium)	OpenAI	$1.750	$14.000	85.9%	86.4%	-
GPT-5 (high)	OpenAI	$1.250	$10.000	87.1%	85.4%	112.4
GPT-5 Codex (high)	OpenAI	$1.250	$10.000	86.5%	83.7%	156.2
Claude Opus 4.5 (Non-reasoning)	Anthropic	$5.000	$25.000	88.9%	81.0%	79.0
Claude 4.5 Sonnet (Reasoning)	Anthropic	$3.000	$15.000	87.5%	83.4%	77.6
GPT-5 (medium)	OpenAI	$1.250	$10.000	86.7%	84.2%	103.1
GPT-5.1 Codex (high)	OpenAI	$1.250	$10.000	86.0%	86.0%	202.5
GLM-4.7 (Reasoning)	Z AI	$0.550	$2.150	85.6%	85.9%	123.4
DeepSeek V3.2 (Reasoning)	DeepSeek	$0.280	$0.420	86.2%	84.0%	29.3
Grok 4	xAI	$3.000	$15.000	86.6%	87.7%	39.7
GPT-5 mini (high)	OpenAI	$0.250	$2.000	83.7%	82.8%	75.5
Gemini 3 Pro Preview (low)	Google	$2.000	$12.000	89.5%	88.7%	126.5
o3	OpenAI	$2.000	$8.000	85.3%	82.7%	217.4
o3-pro	OpenAI	$20.000	$80.000	—	84.5%	29.4
Kimi K2 Thinking	Kimi	$0.600	$2.500	84.8%	83.8%	81.2
MiniMax-M2.1	MiniMax	$0.300	$1.200	87.5%	83.0%	70.4
MiMo-V2-Flash (Reasoning)	Xiaomi	$0.100	$0.300	84.3%	84.6%	102.1
GPT-5 mini (medium)	OpenAI	$0.250	$2.000	82.8%	80.3%	74.8
GPT-5 (low)	OpenAI	$1.250	$10.000	86.0%	80.8%	98.9
Claude 4 Sonnet (Reasoning)	Anthropic	$3.000	$15.000	84.2%	77.7%	78.9
GPT-5.1 Codex mini (high)	OpenAI	$0.250	$2.000	82.0%	81.3%	116.5
Grok 4.1 Fast (Reasoning)	xAI	$0.200	$0.500	85.4%	85.3%	121.1
Claude 4.5 Sonnet (Non-reasoning)	Anthropic	$3.000	$15.000	86.0%	72.7%	54.1
Claude 4.5 Haiku (Reasoning)	Anthropic	$1.000	$5.000	76.0%	67.2%	101.2
KAT-Coder-Pro V1	KwaiKAT	$0.300	$1.200	81.3%	76.4%	67.0
MiniMax-M2	MiniMax	$0.300	$1.200	82.0%	77.7%	85.6
Nova 2.0 Pro Preview (medium)	Amazon	$1.250	$10.000	83.0%	78.5%	133.7
Gemini 3 Flash Preview (Non-reasoning)	Google	$0.500	$3.000	88.2%	81.2%	167.8
Grok 4 Fast (Reasoning)	xAI	$0.200	$0.500	85.0%	84.7%	137.7
Claude 3.7 Sonnet (Reasoning)	Anthropic	$3.000	$15.000	83.7%	77.2%	-
Gemini 2.5 Pro	Google	$1.250	$10.000	86.2%	84.4%	152.1
GLM-4.7 (Non-reasoning)	Z AI	$0.550	$2.150	79.4%	66.4%	116.0
DeepSeek V3.1 Terminus (Reasoning)	DeepSeek	$0.400	$2.000	85.1%	79.2%	-
Doubao Seed Code	ByteDance Seed	$0.170	$1.120	85.4%	76.4%	37.0
GPT-5.2 (Non-reasoning)	OpenAI	$1.750	$14.000	81.4%	71.2%	70.1
gpt-oss-120B (high)	OpenAI	$0.150	$0.600	80.8%	78.2%	327.4
o4-mini (high)	OpenAI	$1.100	$4.400	83.2%	78.4%	132.1
Claude 4 Sonnet (Non-reasoning)	Anthropic	$3.000	$15.000	83.7%	68.3%	72.5
DeepSeek V3.2 Exp (Reasoning)	DeepSeek	$0.280	$0.420	85.0%	79.7%	29.1
Grok 3 mini Reasoning (high)	xAI	$0.300	$0.500	82.8%	79.1%	193.2
GLM-4.6 (Reasoning)	Z AI	$0.575	$2.200	82.9%	78.0%	87.6
Qwen3 Max Thinking	Alibaba	$1.200	$6.000	82.4%	77.6%	36.8
Nova 2.0 Pro Preview (low)	Amazon	$1.250	$10.000	82.2%	75.1%	131.0
DeepSeek V3.2 (Non-reasoning)	DeepSeek	$0.280	$0.420	83.7%	75.1%	28.8
Claude 4.1 Opus (Reasoning)	Anthropic	$15.000	$75.000	88.0%	80.9%	48.4
Qwen3 Max	Alibaba	$1.200	$6.000	84.1%	76.4%	32.9
Gemini 2.5 Flash Preview (Sep '25) (Reasoning)	Google	$0.300	$2.500	84.2%	79.3%	290.0
Claude 3.7 Sonnet (Non-reasoning)	Anthropic	$3.000	$15.000	80.3%	65.6%	-
Claude 4.5 Haiku (Non-reasoning)	Anthropic	$1.000	$5.000	80.0%	64.6%	92.4
o1	OpenAI	$15.000	$60.000	84.1%	74.7%	169.7
MiMo-V2-Flash (Non-reasoning)	Xiaomi	$0.100	$0.300	74.4%	65.6%	92.0
Gemini 2.5 Pro Preview (Mar' 25)	Google	$1.250	$10.000	85.8%	83.6%	-
GLM-4.6 (Non-reasoning)	Z AI	$0.600	$2.200	78.4%	63.2%	87.2
Nova 2.0 Lite (medium)	Amazon	$0.300	$2.500	81.3%	76.8%	240.2
Gemini 2.5 Pro Preview (May' 25)	Google	$1.250	$10.000	83.7%	82.2%	-
Qwen3 235B A22B 2507 (Reasoning)	Alibaba	$0.700	$8.400	84.3%	79.0%	75.9
DeepSeek V3.2 Speciale	DeepSeek	$0.400	$0.500	86.3%	87.1%	-
ERNIE 5.0 Thinking Preview	Baidu	$0.840	$3.370	83.0%	77.7%	25.6
Qwen3 VL 32B (Reasoning)	Alibaba	$0.700	$8.400	81.8%	73.3%	50.5
DeepSeek V3.1 Terminus (Non-reasoning)	DeepSeek	$0.400	$1.680	83.6%	75.1%	-
DeepSeek V3.2 Exp (Non-reasoning)	DeepSeek	$0.280	$0.420	83.6%	73.8%	29.4
MiniMax-Text-01	MiniMax	$0.200	$1.100	75.9%	57.8%	28.5
Kimi K2 0905	Kimi	$0.990	$2.500	81.9%	76.7%	47.2
DeepSeek V3.1 (Reasoning)	DeepSeek	$0.590	$1.690	85.1%	77.9%	-
DeepSeek V3.1 (Non-reasoning)	DeepSeek	$0.560	$1.680	83.3%	73.5%	-
Nova 2.0 Omni (medium)	Amazon	$0.300	$2.500	80.9%	76.0%	-
Qwen3 VL 235B A22B (Reasoning)	Alibaba	$0.700	$8.400	83.6%	77.2%	43.6
Magistral Medium 1.2	Mistral	$2.000	$5.000	81.5%	73.9%	35.8
Claude 4 Opus (Reasoning)	Anthropic	$15.000	$75.000	87.3%	79.6%	48.8
Gemini 2.5 Flash (Reasoning)	Google	$0.300	$2.500	83.2%	79.0%	262.3
GPT-5.1 (Non-reasoning)	OpenAI	$1.250	$10.000	80.1%	64.3%	70.4
DeepSeek R1 0528 (May '25)	DeepSeek	$1.350	$4.200	84.9%	81.3%	-
GLM-4.5 (Reasoning)	Z AI	$0.600	$2.200	83.5%	78.2%	50.2
GPT-5 nano (high)	OpenAI	$0.050	$0.400	78.0%	67.6%	115.1
Qwen3 Next 80B A3B (Reasoning)	Alibaba	$0.500	$6.000	82.4%	75.9%	158.4
Grok Code Fast 1	xAI	$0.200	$1.500	79.3%	72.7%	229.5
Qwen3 Max (Preview)	Alibaba	$1.200	$6.000	83.8%	76.4%	32.2
GPT-5 nano (medium)	OpenAI	$0.050	$0.400	77.2%	67.0%	126.3
o3-mini	OpenAI	$1.100	$4.400	79.1%	74.8%	142.7
Kimi K2	Kimi	$0.600	$2.500	82.4%	76.6%	37.3
GPT-4.1	OpenAI	$2.000	$8.000	80.6%	66.6%	76.9
o1-pro	OpenAI	$150.000	$600.000	—	—	-
Gemini 2.5 Flash Preview (Sep '25) (Non-reasoning)	Google	$0.300	$2.500	83.6%	76.6%	249.2
Grok 3	xAI	$3.000	$15.000	79.9%	69.3%	41.9
o3-mini (high)	OpenAI	$1.100	$4.400	80.2%	77.3%	159.1
Seed-OSS-36B-Instruct	ByteDance Seed	$0.210	$0.570	81.5%	72.6%	29.4
Nova 2.0 Lite (low)	Amazon	$0.300	$2.500	78.8%	69.8%	250.8
Qwen3 Coder 480B A35B Instruct	Alibaba	$1.500	$7.500	78.8%	61.8%	40.5
NVIDIA Nemotron 3 Nano 30B A3B (Reasoning)	NVIDIA	$0.060	$0.240	79.4%	75.7%	191.4
gpt-oss-20B (high)	OpenAI	$0.070	$0.200	74.8%	68.8%	307.3
Qwen3 235B A22B 2507 Instruct	Alibaba	$0.700	$2.800	82.8%	75.3%	55.4
GPT-5 (minimal)	OpenAI	$1.250	$10.000	80.6%	67.3%	75.0
MiniMax M1 80k	MiniMax	$0.550	$2.200	81.6%	69.7%	-
Nova 2.0 Omni (low)	Amazon	$0.300	$2.500	79.8%	69.9%	-
GLM-4.6V (Reasoning)	Z AI	$0.300	$0.900	79.9%	71.9%	68.0
GLM-4.5-Air	Z AI	$0.200	$1.100	81.5%	73.3%	98.9
gpt-oss-120B (low)	OpenAI	$0.150	$0.595	77.5%	67.2%	299.1
Grok 4.1 Fast (Non-reasoning)	xAI	$0.200	$0.500	74.3%	63.7%	81.1
Nova 2.0 Pro Preview (Non-reasoning)	Amazon	$1.250	$10.000	77.2%	63.6%	159.0
o1-preview	OpenAI	$16.500	$66.000	—	—	-
Claude 4.1 Opus (Non-reasoning)	Anthropic	$15.000	$75.000	—	—	37.9
GPT-4.1 mini	OpenAI	$0.400	$1.600	78.1%	66.4%	67.2
Qwen3 30B A3B 2507 (Reasoning)	Alibaba	$0.200	$2.400	80.5%	70.7%	165.3
Grok 4 Fast (Non-reasoning)	xAI	$0.200	$0.500	73.0%	60.6%	127.6
DeepSeek V3 0324	DeepSeek	$1.195	$1.350	81.9%	65.5%	-
Ring-1T	InclusionAI	$0.560	$2.240	80.6%	77.4%	57.9
Mistral Large 3	Mistral	$0.500	$1.500	80.7%	68.0%	50.9
Magistral Small 1.2	Mistral	$0.500	$1.500	76.8%	66.3%	192.7
Gemini 2.5 Flash-Lite Preview (Sep '25) (Reasoning)	Google	$0.100	$0.400	80.8%	70.9%	576.3
INTELLECT-3	Prime Intellect	$0.200	$1.100	82.2%	76.1%	84.8
Claude 4 Opus (Non-reasoning)	Anthropic	$15.000	$75.000	86.0%	70.1%	40.2
CompactifAI Llama 4 Scout Slim	Multiverse Computing	$0.070	$0.100	70.3%	42.6%	116.7
GPT-5 (ChatGPT)	OpenAI	$1.250	$10.000	82.0%	68.6%	169.6
Hermes 4 - Llama-3.1 405B (Reasoning)	Nous Research	$1.000	$3.000	82.9%	72.7%	33.8
GPT-5 mini (minimal)	OpenAI	$0.250	$2.000	77.5%	68.7%	72.1
gpt-oss-20B (low)	OpenAI	$0.070	$0.200	71.8%	61.1%	246.9
Mistral Medium 3.1	Mistral	$0.400	$2.000	68.3%	58.8%	90.6
Gemini 2.5 Flash (Non-reasoning)	Google	$0.300	$2.500	80.9%	68.3%	224.2
Qwen3 VL 235B A22B Instruct	Alibaba	$0.700	$2.800	82.3%	71.2%	35.3
Ring-flash-2.0	InclusionAI	$0.140	$0.570	79.3%	72.5%	87.3
Hermes 4 - Llama-3.1 70B (Reasoning)	Nous Research	$0.130	$0.400	81.1%	69.9%	81.0
Qwen3 Coder 30B A3B Instruct	Alibaba	$0.450	$2.250	70.6%	51.6%	98.6
Qwen3 Next 80B A3B Instruct	Alibaba	$0.500	$2.000	81.9%	73.8%	145.9
Codestral (Jan '25)	Mistral	$0.300	$0.900	44.6%	31.2%	212.0
Gemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning)	Google	$0.100	$0.400	79.6%	65.1%	450.6
Qwen3 235B A22B (Reasoning)	Alibaba	$0.700	$8.400	82.8%	70.0%	54.2
Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)	NVIDIA	$0.600	$1.800	82.5%	72.8%	37.2
Ling-flash-2.0	InclusionAI	$0.140	$0.570	77.7%	65.7%	54.6
Qwen3 VL 30B A3B (Reasoning)	Alibaba	$0.200	$2.400	80.7%	72.0%	106.6
QwQ 32B	Alibaba	$0.430	$0.600	76.4%	59.3%	29.6
Llama Nemotron Super 49B v1.5 (Reasoning)	NVIDIA	$0.100	$0.400	81.4%	74.8%	74.1
Reka Core	Reka AI	$2.000	$2.000	—	—	40.9
GLM-4.5V (Reasoning)	Z AI	$0.600	$1.800	78.8%	68.4%	31.4
Nova Premier	Amazon	$2.500	$12.500	73.3%	56.9%	75.1
Magistral Medium 1	Mistral	$2.000	$5.000	75.3%	67.9%	60.2
GPT-4o (Aug '24)	OpenAI	$2.500	$10.000	—	52.1%	79.0
Llama 4 Maverick	Meta	$0.310	$0.850	80.9%	67.1%	127.7
Gemini 2.0 Flash (Feb '25)	Google	$0.100	$0.400	77.9%	62.3%	-
Devstral Medium	Mistral	$0.400	$2.000	70.8%	49.2%	112.4
Nova 2.0 Lite (Non-reasoning)	Amazon	$0.300	$2.500	74.3%	60.3%	224.9
Claude 3.5 Haiku	Anthropic	$0.800	$4.000	63.4%	40.8%	48.4
DeepSeek R1 (Jan '25)	DeepSeek	$1.350	$4.000	84.4%	70.8%	-
Reka Flash (Feb '24)	Reka AI	$0.200	$0.800	—	—	69.1
GPT-4o (March 2025, chatgpt-4o-latest)	OpenAI	$5.000	$15.000	80.3%	65.5%	215.5
Reka Edge	Reka AI	$0.100	$0.100	—	—	63.8
Gemini 2.5 Flash-Lite (Reasoning)	Google	$0.100	$0.400	75.9%	62.5%	369.9
Sonar Reasoning	Perplexity	$1.000	$5.000	—	62.3%	-
Devstral Small (May '25)	Mistral	$0.100	$0.300	63.2%	43.4%	200.0
Llama 3.1 Instruct 405B	Meta	$3.750	$6.750	73.2%	51.5%	24.9
Mistral Medium 3	Mistral	$0.400	$2.000	76.0%	57.8%	95.7
GLM-4.6V (Non-reasoning)	Z AI	$0.300	$0.900	75.2%	56.6%	46.6
Nova 2.0 Omni (Non-reasoning)	Amazon	$0.300	$2.500	71.9%	55.5%	226.3
Qwen3 235B A22B (Non-reasoning)	Alibaba	$0.700	$2.800	76.2%	61.3%	45.7
ERNIE 4.5 300B A47B	Baidu	$0.280	$1.100	77.6%	81.1%	26.4
Qwen3 VL 32B Instruct	Alibaba	$0.700	$2.800	79.1%	67.1%	45.6
DeepSeek R1 Distill Qwen 32B	DeepSeek	$0.285	$0.285	73.9%	61.5%	39.4
Qwen3 VL 8B (Reasoning)	Alibaba	$0.180	$2.100	74.9%	57.9%	63.6
Qwen3 32B (Reasoning)	Alibaba	$0.700	$8.400	79.8%	66.8%	88.7
DeepSeek V3 (Dec '24)	DeepSeek	$0.400	$0.890	75.2%	55.7%	-
Hermes 4 - Llama-3.1 405B (Non-reasoning)	Nous Research	$1.000	$3.000	72.9%	53.6%	32.9
EXAONE 4.0 32B (Reasoning)	LG AI Research	$0.600	$1.000	81.8%	73.9%	96.2
Qwen3 14B (Reasoning)	Alibaba	$0.350	$4.200	77.4%	60.4%	58.8
Magistral Small 1	Mistral	$0.500	$1.500	74.6%	64.1%	118.7
Olmo 3 7B Think	Allen Institute for AI	$0.120	$0.200	65.5%	51.6%	170.3
CompactifAI Llama 3.3 70B Slim	Multiverse Computing	$0.160	$0.310	57.1%	35.5%	130.7
CompactifAI Mistral Small 3.1 Slim	Multiverse Computing	$0.050	$0.080	53.8%	32.9%	121.6
Qwen3 VL 30B A3B Instruct	Alibaba	$0.200	$0.800	76.4%	69.5%	99.2
DeepSeek R1 0528 Qwen3 8B	DeepSeek	$0.060	$0.090	73.9%	61.2%	-
Ministral 3 14B	Mistral	$0.200	$0.200	69.3%	57.2%	126.2
Qwen2.5 Max	Alibaba	$1.600	$6.400	76.2%	58.7%	41.4
DeepSeek R1 Distill Llama 70B	DeepSeek	$0.875	$1.300	79.5%	40.2%	38.9
Claude 3.5 Sonnet (Oct '24)	Anthropic	$3.000	$15.000	77.2%	59.9%	-
DeepSeek R1 Distill Qwen 14B	DeepSeek	$0.150	$0.150	74.0%	48.4%	-
Qwen3 30B A3B (Reasoning)	Alibaba	$0.200	$2.400	77.7%	61.6%	66.5
Qwen3 Omni 30B A3B (Reasoning)	Alibaba	$0.250	$0.970	79.2%	72.6%	97.1
Devstral Small (Jul '25)	Mistral	$0.100	$0.300	62.2%	41.4%	237.5
Qwen3 30B A3B 2507 Instruct	Alibaba	$0.200	$0.800	77.7%	65.9%	59.8
Llama Nemotron Super 49B v1.5 (Non-reasoning)	NVIDIA	$0.100	$0.400	69.2%	48.1%	69.4
NVIDIA Nemotron Nano 9B V2 (Reasoning)	NVIDIA	$0.040	$0.160	74.2%	57.0%	118.2
Sonar	Perplexity	$1.000	$1.000	68.9%	47.1%	78.3
Mistral Small 3.2	Mistral	$0.100	$0.300	68.1%	50.5%	111.3
Qwen3 8B (Reasoning)	Alibaba	$0.180	$2.100	74.3%	58.9%	85.1
Ministral 3 8B	Mistral	$0.150	$0.150	64.2%	47.1%	196.9
NVIDIA Nemotron Nano 12B v2 VL (Reasoning)	NVIDIA	$0.200	$0.600	75.9%	57.2%	128.9
QwQ 32B-Preview	Alibaba	$0.120	$0.180	64.8%	55.7%	40.8
Sonar Pro	Perplexity	$3.000	$15.000	75.5%	57.8%	86.2
Llama 3.3 Instruct 70B	Meta	$0.585	$0.715	71.3%	49.8%	105.5
Ling-mini-2.0	InclusionAI	$0.070	$0.280	67.1%	56.2%	156.9
GPT-4o (Nov '24)	OpenAI	$2.500	$10.000	74.8%	54.3%	116.4
Qwen3 VL 8B Instruct	Alibaba	$0.180	$0.700	68.6%	42.7%	98.3
Gemini 2.0 Flash-Lite (Feb '25)	Google	$0.075	$0.300	72.4%	53.5%	-
Mistral Large 2 (Nov '24)	Mistral	$2.000	$6.000	69.7%	48.6%	45.9
Gemini 2.0 Flash-Lite (Preview)	Google	$0.075	$0.300	—	54.2%	-
GPT-4o (May '24)	OpenAI	$5.000	$15.000	74.0%	52.6%	78.4
Qwen3 32B (Non-reasoning)	Alibaba	$0.700	$2.800	72.7%	53.5%	85.6
Phi-3 Medium Instruct 14B	Microsoft Azure	$0.170	$0.680	54.3%	32.6%	-
Reka Flash 3	Reka AI	$0.200	$0.800	66.9%	52.9%	49.3
Claude 3.5 Sonnet (June '24)	Anthropic	$3.000	$15.000	75.1%	56.0%	-
Qwen3 4B (Reasoning)	Alibaba	$0.110	$1.260	69.6%	52.2%	91.0
Llama 3.1 Nemotron Instruct 70B	NVIDIA	$1.200	$1.200	69.0%	46.5%	39.6
GPT-4o (ChatGPT)	OpenAI	$5.000	$15.000	77.3%	51.1%	202.7
GPT-5 nano (minimal)	OpenAI	$0.050	$0.400	55.6%	42.8%	116.8
Llama 4 Scout	Meta	$0.180	$0.625	75.2%	58.7%	141.7
NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning)	NVIDIA	$0.060	$0.240	57.9%	39.9%	174.9
Mistral Small 3.1	Mistral	$0.100	$0.300	65.9%	45.4%	106.0
Pixtral Large	Mistral	$2.000	$6.000	70.1%	50.5%	30.6
NVIDIA Nemotron Nano 9B V2 (Non-reasoning)	NVIDIA	$0.060	$0.230	73.9%	55.7%	111.8
Command A	Cohere	$2.500	$10.000	71.2%	52.7%	52.8
GPT-4 Turbo	OpenAI	$10.000	$30.000	69.4%	—	27.0
Hermes 4 - Llama-3.1 70B (Non-reasoning)	Nous Research	$0.130	$0.400	66.4%	49.1%	73.0
Aya Expanse 32B	Cohere	$0.500	$1.500	37.7%	23.0%	41.9
Nova Pro	Amazon	$0.800	$3.200	69.1%	49.9%	-
Qwen3 14B (Non-reasoning)	Alibaba	$0.350	$1.400	67.5%	47.0%	54.8
GPT-4.1 nano	OpenAI	$0.100	$0.400	65.7%	51.2%	130.1
Phi-4	Microsoft Azure	$0.125	$0.500	71.4%	57.5%	10.9
GLM-4.5V (Non-reasoning)	Z AI	$0.600	$1.800	75.1%	57.3%	28.3
Gemini 2.5 Flash-Lite (Non-reasoning)	Google	$0.100	$0.400	72.4%	47.4%	256.1
Llama 3.1 Instruct 70B	Meta	$0.560	$0.560	67.6%	40.9%	61.5
Qwen3 1.7B (Reasoning)	Alibaba	$0.110	$1.260	57.0%	35.6%	124.8
Mistral Large 2 (Jul '24)	Mistral	$2.000	$6.000	68.3%	47.2%	-
Qwen2.5 Coder Instruct 32B	Alibaba	$0.130	$0.175	63.5%	41.7%	35.7
GPT-4	OpenAI	$30.000	$60.000	—	—	28.3
Mistral Small 3	Mistral	$0.100	$0.300	65.2%	46.2%	231.4
GPT-4o mini	OpenAI	$0.150	$0.600	64.8%	42.6%	49.0
Qwen3 4B (Non-reasoning)	Alibaba	$0.110	$0.420	58.6%	39.8%	85.1
Claude 3 Opus	Anthropic	$15.000	$75.000	69.6%	48.9%	-
Nova Lite	Amazon	$0.060	$0.240	59.0%	43.3%	238.3
Qwen3 30B A3B (Non-reasoning)	Alibaba	$0.200	$0.800	71.0%	51.5%	61.7
Ministral 8B	Mistral	$0.100	$0.100	38.9%	27.6%	195.9
Llama 3.1 Instruct 8B	Meta	$0.100	$0.100	47.6%	25.9%	162.8
Ministral 3 3B	Mistral	$0.100	$0.100	52.4%	35.8%	293.0
Olmo 3.1 32B Instruct	Allen Institute for AI	$0.200	$0.600	—	53.9%	47.6
Reka Flash (Sep '24)	Reka AI	$0.200	$0.800	—	—	71.1
EXAONE 4.0 32B (Non-reasoning)	LG AI Research	$0.600	$1.000	76.8%	62.8%	88.3
Qwen2.5 Turbo	Alibaba	$0.050	$0.200	63.3%	41.0%	66.4
Llama 3.2 Instruct 90B (Vision)	Meta	$0.720	$0.720	67.1%	43.2%	36.6
Solar Mini	Upstage	$0.150	$0.150	—	—	84.5
Granite 4.0 H Small	IBM	$0.060	$0.250	62.4%	41.6%	370.8
CompactifAI Llama 3.1 8B Slim	Multiverse Computing	$0.050	$0.070	32.1%	22.1%	228.4
Llama 3.2 Instruct 11B (Vision)	Meta	$0.160	$0.160	46.4%	22.1%	48.8
Ministral 3B	Mistral	$0.040	$0.040	33.9%	26.0%	272.6
Granite 3.3 8B (Non-reasoning)	IBM	$0.030	$0.250	46.8%	33.8%	488.2
Qwen3 8B (Non-reasoning)	Alibaba	$0.180	$0.700	64.3%	45.2%	79.7
Qwen3 Omni 30B A3B Instruct	Alibaba	$0.250	$0.970	72.5%	62.0%	88.3
Jamba 1.5 Large	AI21 Labs	$2.000	$8.000	57.2%	42.7%	-
Jamba 1.6 Large	AI21 Labs	$2.000	$8.000	56.5%	38.7%	44.9
Hermes 3 - Llama-3.1 70B	Nous Research	$0.300	$0.300	57.1%	40.1%	41.3
NVIDIA Nemotron Nano 12B v2 VL (Non-reasoning)	NVIDIA	$0.200	$0.600	64.9%	43.9%	131.2
Nova Micro	Amazon	$0.035	$0.140	53.1%	35.8%	421.4
Claude 3 Sonnet	Anthropic	$3.000	$15.000	57.9%	40.0%	-
Mistral Small (Sep '24)	Mistral	$0.200	$0.600	52.9%	38.1%	110.9
Llama 3 Instruct 70B	Meta	$0.510	$0.740	57.4%	37.9%	38.0
Phi-3 Mini Instruct 3.8B	Microsoft Azure	$0.130	$0.520	43.5%	31.9%	-
Aya Expanse 8B	Cohere	$0.500	$1.500	31.2%	24.7%	79.5
Mistral Large (Feb '24)	Mistral	$4.000	$12.000	51.5%	35.1%	-
Llama 3.2 Instruct 3B	Meta	$0.060	$0.060	34.7%	25.5%	46.5
Llama 2 Chat 7B	Meta	$0.050	$0.250	16.4%	22.7%	109.5
Jamba 1.7 Large	AI21 Labs	$2.000	$8.000	57.7%	39.0%	41.9
Claude 3 Haiku	Anthropic	$0.250	$1.250	—	—	111.8
Llama 3.2 Instruct 1B	Meta	$0.053	$0.055	20.0%	19.6%	74.5
Mistral Small (Feb '24)	Mistral	$1.000	$3.000	41.9%	30.2%	108.1
GPT-3.5 Turbo	OpenAI	$0.500	$1.500	46.2%	29.7%	112.3
Mistral Medium	Mistral	$2.750	$8.100	49.1%	34.9%	92.0
Pixtral 12B (2409)	Mistral	$0.150	$0.150	47.3%	34.3%	136.6
Llama 3 Instruct 8B	Meta	$0.045	$0.155	40.5%	29.6%	67.1
Command-R+ (Apr '24)	Cohere	$3.000	$15.000	43.2%	32.3%	-
Olmo 3 7B Instruct	Allen Institute for AI	$0.100	$0.200	52.2%	40.0%	37.0
Jamba 1.5 Mini	AI21 Labs	$0.200	$0.400	37.1%	30.2%	-
Jamba 1.6 Mini	AI21 Labs	$0.200	$0.400	36.7%	30.0%	127.1
Jamba 1.7 Mini	AI21 Labs	$0.200	$0.400	38.8%	32.2%	158.1
Gemma 2 9B	Google	$0.030	$0.090	49.5%	31.1%	-
Mixtral 8x7B Instruct	Mistral	$0.540	$0.600	38.7%	29.2%	-
Command-R (Mar '24)	Cohere	$0.500	$1.500	33.8%	28.4%	-
Mistral 7B Instruct	Mistral	$0.250	$0.250	24.5%	17.7%	125.6
Command-R+ (Aug '24)	Cohere	$2.500	$10.000	42.7%	33.7%	21.0
Qwen3 1.7B (Non-reasoning)	Alibaba	$0.110	$0.420	41.1%	28.3%	116.5
Qwen3 0.6B (Reasoning)	Alibaba	$0.110	$1.260	34.7%	23.9%	199.0
Gemma 3n E4B Instruct	Google	$0.020	$0.040	48.8%	29.6%	60.5
Qwen3 0.6B (Non-reasoning)	Alibaba	$0.110	$0.420	23.1%	23.1%	190.1
Mistral NeMo	Mistral	$0.150	$0.150	39.9%	31.4%	190.2
Command-R (Aug '24)	Cohere	$0.150	$0.600	33.7%	28.9%	59.8
Cogito v2.1 (Reasoning)	Deep Cogito	$1.250	$1.250	84.9%	76.8%	73.9
DeepSeek-OCR	DeepSeek	$0.030	$0.100	—	—	306.8
Grok 3 mini Reasoning (Low)	xAI	$0.300	$0.500	—	—	110.8
Doubao-Seed-1.8	ByteDance Seed	$0.110	$0.280	—	—	43.1

AI Stats

No models found