AI Stats | AI Model Pricing and Benchmarks

Tracked models 376

Priced models from the Artificial Analysis feed.

AA average leader GPT-5 (high) 69.6% avg

Cheapest blended Qwen3.5 0.8B (Reasoning) $0.020

Epoch signal GPT-5.4 Pro 74.1% avg

Epoch coverage 611

49 benchmarks / 3447 benchmark runs

Token averages

Input $1.866 Output $7.918

Mean price per 1M tokens across priced models.

Public catalog sources

Public catalog data

Rank by

OR models: 345
OR routes: 363
Providers: 83
Embeddings: 26
HF matched: 91
LiteLLM matched: 246
OR matched: 249
OR usage: 0
Route detail: 39

Public metadata from OpenRouter, Hugging Face, and LiteLLM adds availability, provider directory data, embeddings, usage, context, route pricing, endpoint coverage, and capability checks. Benchmark scores remain AA and Epoch.

Provider coverage OpenRouter route catalog ranking

Leaderboard

Switch metrics to see different top-5 rankings.

Metric

Source: Artificial Analysis

All models

Search and scan every tracked model.

Showing 8 of 376 models

Search models

Sort models

Claude Opus 4.8 (Adaptive Reasoning, Max Effort)

Anthropic

AA + 2 catalogs

Pricing

Input (1M): $6.250
Output (1M): $25.000
Blended (3:1): $10.938

Performance

Tokens / s: 55.1
TTFT (token): 16.62s
TTFT (answer): 16.62s

Specs

Context: 1M
Max output: -
Source: AA + 2 catalogs

Public catalog

Tools: Yes
Route: Paid
OR input: $10.000
OR output: $50.000
HF signal: -
OR usage: -
OR routes: -
Context: 1M
Capabilities: Tools, Vision, Reasoning, Cache

Benchmarks

MMLU Pro N/A

GPQA 92.0%

HLE 45.7%

AIME N/A

LiveCodeBench N/A

SciCode 53.5%

Math 500 N/A

AA Indexes

Intelligence

61.4%

Coding

56.7%

Math

N/A

GPT-5.5 (xhigh)

OpenAI

AA + 3 catalogs

Pricing

Input (1M): $5.000
Output (1M): $30.000
Blended (3:1): $11.250

Performance

Tokens / s: 70.7
TTFT (token): 43.39s
TTFT (answer): 43.39s

Specs

Context: 1.1M
Max output: -
Source: AA + 3 catalogs

Public catalog

Tools: Yes
Route: Paid
HF signal: -
OR usage: -
OR routes: 2 providers
Context: 1.1M
Capabilities: Tools, Vision, Reasoning, Cache, Web

Benchmarks

MMLU Pro N/A

GPQA 93.5%

HLE 44.3%

AIME N/A

LiveCodeBench N/A

SciCode 56.1%

Math 500 N/A

AA Indexes

Intelligence

60.2%

Coding

59.1%

Math

N/A

GPT-5.5 (high)

OpenAI

AA + 3 catalogs

Pricing

Input (1M): $5.000
Output (1M): $30.000
Blended (3:1): $11.250

Performance

Tokens / s: 61.2
TTFT (token): 17.08s
TTFT (answer): 17.08s

Specs

Context: 1.1M
Max output: -
Source: AA + 3 catalogs

Public catalog

Tools: Yes
Route: Paid
HF signal: -
OR usage: -
OR routes: 2 providers
Context: 1.1M
Capabilities: Tools, Vision, Reasoning, Cache, Web

Benchmarks

MMLU Pro N/A

GPQA 93.2%

HLE 43.0%

AIME N/A

LiveCodeBench N/A

SciCode 55.9%

Math 500 N/A

AA Indexes

Intelligence

58.9%

Coding

58.5%

Math

N/A

Claude Opus 4.7 (Adaptive Reasoning, Max Effort)

Anthropic

AA + 2 catalogs

Pricing

Input (1M): $6.250
Output (1M): $25.000
Blended (3:1): $10.938

Performance

Tokens / s: 49.5
TTFT (token): 14.25s
TTFT (answer): 14.25s

Specs

Context: 1M
Max output: -
Source: AA + 2 catalogs

Public catalog

Tools: Yes
Route: Paid
OR input: $30.000
OR output: $150.000
HF signal: -
OR usage: -
OR routes: -
Context: 1M
Capabilities: Tools, Vision, Reasoning, Cache

Benchmarks

MMLU Pro N/A

GPQA 91.4%

HLE 39.6%

AIME N/A

LiveCodeBench N/A

SciCode 54.5%

Math 500 N/A

AA Indexes

Intelligence

57.3%

Coding

52.5%

Math

N/A

Gemini 3.1 Pro Preview

Google

AA + 3 catalogs

Pricing

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

Performance

Tokens / s: 120.2
TTFT (token): 22.54s
TTFT (answer): 22.54s

Specs

Context: 1.0M
Max output: -
Source: AA + 3 catalogs

Public catalog

Tools: Yes
Route: Paid
HF signal: -
OR usage: -
OR routes: 2 providers
Context: 1.0M
Capabilities: Tools, Vision, Reasoning, Cache, Web

Benchmarks

MMLU Pro N/A

GPQA 94.1%

HLE 44.7%

AIME N/A

LiveCodeBench N/A

SciCode 58.9%

Math 500 N/A

AA Indexes

Intelligence

57.2%

Coding

55.5%

Math

N/A

GPT-5.4 (xhigh)

OpenAI

AA + 3 catalogs

Pricing

Input (1M): $2.500
Output (1M): $15.000
Blended (3:1): $5.625

Performance

Tokens / s: 78.5
TTFT (token): 198.50s
TTFT (answer): 198.50s

Specs

Context: 1.1M
Max output: -
Source: AA + 3 catalogs

Public catalog

Tools: Yes
Route: Paid
HF signal: -
OR usage: -
OR routes: 2 providers
Context: 1.1M
Capabilities: Tools, Vision, Reasoning, Cache, Web

Benchmarks

MMLU Pro N/A

GPQA 92.0%

HLE 41.6%

AIME N/A

LiveCodeBench N/A

SciCode 56.6%

Math 500 N/A

AA Indexes

Intelligence

56.8%

Coding

57.2%

Math

N/A

GPT-5.5 (medium)

OpenAI

AA + 3 catalogs

Pricing

Input (1M): $5.000
Output (1M): $30.000
Blended (3:1): $11.250

Performance

Tokens / s: 55.0
TTFT (token): 5.23s
TTFT (answer): 5.23s

Specs

Context: 1.1M
Max output: -
Source: AA + 3 catalogs

Public catalog

Tools: Yes
Route: Paid
HF signal: -
OR usage: -
OR routes: 2 providers
Context: 1.1M
Capabilities: Tools, Vision, Reasoning, Cache, Web

Benchmarks

MMLU Pro N/A

GPQA 92.6%

HLE 40.6%

AIME N/A

LiveCodeBench N/A

SciCode 53.5%

Math 500 N/A

AA Indexes

Intelligence

56.7%

Coding

56.2%

Math

N/A

Qwen3.7 Max

Alibaba

AA + OR

Pricing

Input (1M): $2.500
Output (1M): $7.500
Blended (3:1): $3.750

Performance

Tokens / s: 186.5
TTFT (token): 1.75s
TTFT (answer): 14.67s

Specs

Context: 1M
Max output: -
Source: AA + OR

Public catalog

Tools: Yes
Route: Paid
OR input: $1.250
OR output: $3.750
HF signal: -
OR usage: -
OR routes: -
Context: 1M
Capabilities: Tools

Benchmarks

MMLU Pro N/A

GPQA 92.3%

HLE 38.1%

AIME N/A

LiveCodeBench N/A

SciCode 48.8%

Math 500 N/A

AA Indexes

Intelligence

56.6%

Coding

50.1%

Math

N/A

Model	Provider	Context	Input	Output	MMLU	GPQA	TPS
Claude Opus 4.8 (Adaptive Reasoning, Max Effort)	Anthropic	1M	$6.250	$25.000	N/A	92.0%	55.1
GPT-5.5 (xhigh)	OpenAI	1.1M	$5.000	$30.000	N/A	93.5%	70.7
GPT-5.5 (high)	OpenAI	1.1M	$5.000	$30.000	N/A	93.2%	61.2
Claude Opus 4.7 (Adaptive Reasoning, Max Effort)	Anthropic	1M	$6.250	$25.000	N/A	91.4%	49.5
Gemini 3.1 Pro Preview	Google	1.0M	$2.000	$12.000	N/A	94.1%	120.2
GPT-5.4 (xhigh)	OpenAI	1.1M	$2.500	$15.000	N/A	92.0%	78.5
GPT-5.5 (medium)	OpenAI	1.1M	$5.000	$30.000	N/A	92.6%	55.0
Qwen3.7 Max	Alibaba	1M	$2.500	$7.500	N/A	92.3%	186.5

AI model pricing and benchmark dashboard

No models found