Best AI models for knowledge
MMLU Pro tests broad knowledge across academic and professional subjects with harder, reasoning-heavy questions than the original MMLU. A strong score indicates wide, reliable factual coverage.
- 馃
gemini-3-pro-previewGoogle Gemini API路$2.00 / $12.00 per 1M89.8%89.8% - 馃
claude-opus-4-5Anthropic路$5.00 / $25.00 per 1M89.5%89.5% - 馃gemini-3-flash-previewGoogle Gemini API路$0.50 / $3.00 per 1M89.0%89.0%
- 4
claude-opus-4-1Google Vertex AI路$15.00 / $75.00 per 1M88.0%88.0% - 5claude-sonnet-4-5Anthropic路$3.00 / $15.00 per 1M87.5%87.5%
- 6
gpt-5.2OpenAI路$1.75 / $14.00 per 1M87.4%87.4% - 7claude-opus-4Google Vertex AI路$15.00 / $75.00 per 1M87.3%87.3%
- 8gpt-5OpenAI路$1.25 / $10.00 per 1M87.1%87.1%
- 9gpt-5.1OpenAI路$1.25 / $10.00 per 1M87.0%87.0%
- 10grok-4xAI路$3.00 / $15.00 per 1M86.6%86.6%
- 11deepseek-v3.2Google Vertex AI路$0.56 / $1.68 per 1M86.2%86.2%
- 12gemini-2.5-proGoogle Gemini API路$1.25 / $10.00 per 1M86.2%86.2%
- 13
GLM-4.7Z.ai路$0.60 / $2.20 per 1M85.6%85.6% - 14o3OpenAI路$2.00 / $8.00 per 1M85.3%85.3%
- 15grok-4-fastxAI路$0.20 / $0.50 per 1M85.0%85.0%
- 16
deepseek-r1Novita AI路$4.00 / $4.00 per 1M84.9%84.9% - 17kimi-k2Google Vertex AI路$0.60 / $2.50 per 1M84.8%84.8%
- 18
Qwen/Qwen3-235B-A22B-Instruct-2507DeepInfra路$0.07 / $0.10 per 1M84.3%84.3% - 19claude-sonnet-4Google Vertex AI路$3.00 / $15.00 per 1M84.2%84.2%
- 20o1OpenAI路$15.00 / $60.00 per 1M84.1%84.1%
- 21deepseek-ai/DeepSeek-V3.1DeepInfra路$0.30 / $1.00 per 1M83.3%83.3%
- 22gemini-2.5-flashGoogle Gemini API路$0.30 / $2.50 per 1M83.2%83.2%
- 23o4-miniOpenAI路$1.10 / $4.40 per 1M83.2%83.2%
- 24GLM-4.6Z.ai路$0.60 / $2.20 per 1M82.9%82.9%
- 25gpt-5-miniOpenAI路$0.25 / $2.00 per 1M82.8%82.8%
- 26Qwen/Qwen3-235B-A22BDeepInfra路$0.20 / $0.60 per 1M82.8%82.8%
- 27grok-3-minixAI路$0.30 / $0.50 per 1M82.8%82.8%
- 28minimax-m2MiniMax路$0.30 / $1.20 per 1M82.0%82.0%
- 29deepseek-v3-0324Novita AI路$0.40 / $1.30 per 1M81.9%81.9%
- 30zai-org/GLM-4.5-AirDeepInfra路$0.20 / $1.10 per 1M81.5%81.5%
Explore other rankings
Smartest overall
Ranked by Intelligence Index
Best for coding
Ranked by Coding Index
Best coding agent
Ranked by Terminal-Bench Hard
Best for reasoning
Ranked by GPQA Diamond
Best at math
Ranked by Math Index
Best for tool use
Ranked by 蟿虏-Bench
Cheapest
Lowest input + output price per 1M tokens
Longest context
Max tokens in a single prompt
How we rank
Scores for MMLU Pro come from Artificial Analysis, an independent AI benchmarking service. When a model is available through multiple providers (e.g. Anthropic direct, AWS Bedrock, Google Vertex), we show one canonical entry per model family so the ranking isn't polluted by duplicates. Benchmarks measure specific skills, so always validate on your own workload before committing.
One API for every model on this list
Requesty is OpenAI-compatible and routes to 600+ models. Switch between any of the models above by changing one parameter in your code.