AnotherWrapper
    Build with the APIs
    LLMImageVideoSpeechTranscription

    HumanEval

    🏆 Leaderboard

    As of August 31, 2026, MiniCPM-SALA is #1 for HumanEval at 95.1%. Ranked by the HumanEval score 69 models in this index have a published HumanEval score. Methodology: HumanEval (https://github.com/openai/human-eval). HumanEval leaderboard with live API prices. Measures functional code generation — the model writes Python functions from docstrings and is evaluated on unit tests. Official methodology: HumanEval (https://github.com/openai/human-eval).

    Updated August 31, 2026345 models36 providers

    HumanEval vs price

    69 models. Left is cheaper. Up is a higher score. The line is the best score you can buy at each price — a dot under it is a worse deal than something on the line.

    Best score at each price#1 on this board
    MiniCPM-SALAOSS
    OpenBMB · Open Source
    		95.1%$0.20
    Kimi K2 0905
    Moonshot AI · Proprietary
    		94.5%$3.10
    Claude 3.5 Sonnet
    Anthropic · Proprietary
    		93.7%$18.00
    GPT-5
    OpenAI · Proprietary
    		93.4%$11.25
    Kimi K2 InstructOSS
    Moonshot AI · Open Source
    		93.3%$1.00
    Qwen2.5-Coder 32B InstructOSS
    Qwen · Open Source
    		92.7%$0.18
    o1-mini
    OpenAI · Proprietary
    		92.4%$15.00
    Sarvam-30BOSS
    Sarvam AI · Open Source
    		92.1%$0.60
    Mistral Large 2OSS
    Mistral · Open Source · via Mistral AI
    		92%$8.00
    Claude 3.5 Sonnet
    Anthropic · Proprietary
    		92%$18.00
    Qwen2.5 VL 32B InstructOSS
    Qwen · Open Source
    		91.5%$0.90
    GPT-4o
    OpenAI · Proprietary
    		90.2%$12.50
    Granite 3.3 8B InstructOSS
    IBM · Open Source
    		89.7%$1.00
    Gemini Diffusion
    Google · Proprietary
    		89.6%$2.80
    Nova Pro
    Amazon · Proprietary
    		89%$4.00
    Llama 3.1 405B InstructOSS
    Meta · Open Source
    		89%$1.78
    DeepSeek-V2.5OSS
    DeepSeek · Open Source
    		89%$0.42
    Mistral Small 3.1 24B InstructOSS
    Mistral · Open Source · via OpenRouter
    		88.4%$0.91
    LongCat-Flash-ChatOSS
    Meituan · Open Source
    		88.4%$1.50
    Qwen2.5-Coder 7B InstructOSS
    Qwen · Open Source
    		88.4%$0.60
    Qwen2.5 32B InstructOSS
    Qwen · Open Source
    		88.4%$3.50
    Llama 3.3 70B InstructOSS
    Meta · Open Source
    		88.4%$0.40
    Grok-2
    xAI · Proprietary
    		88.4%$12.00
    Gemini 2.5 Pro
    Google · Proprietary
    		88.2%$11.25
    o1
    OpenAI · Proprietary
    		88.1%$75.00
    Claude 3.5 Haiku
    Anthropic · Proprietary
    		88.1%$4.80
    GPT-4.5
    OpenAI · Proprietary
    		88%$225.00
    Gemma 3 27BOSS
    Google · Open Source
    		87.8%$0.30
    GPT-4o mini
    OpenAI · Proprietary
    		87.2%$0.75
    GPT-4 Turbo
    OpenAI · Proprietary
    		87.1%$40.00
    Qwen2.5 72B InstructOSS
    Qwen · Open Source
    		86.6%$0.75
    Qwen2 72B InstructOSS
    Qwen · Open Source
    		86%$0.75
    Nova Lite
    Amazon · Proprietary
    		85.4%$0.30
    Gemma 3 12BOSS
    Google · Open Source
    		85.4%$0.15
    GLM-4.7OSS
    Z AI · Open Source
    		84.9%$2.80
    Claude 3 Opus
    Anthropic · Proprietary
    		84.9%$90.00
    Qwen2.5 7B InstructOSS
    Qwen · Open Source
    		84.8%$0.60
    Mistral Small 3 24B InstructOSS
    Mistral · Open Source · via Mistral AI
    		84.8%$0.21
    Gemini 1.5 Pro
    Google · Proprietary
    		84.1%$12.50
    Qwen2.5 14B InstructOSS
    Qwen · Open Source
    		83.5%$1.75
    Phi 4OSS
    Microsoft · Open Source
    		82.6%$0.21
    IBM Granite 4.0 Tiny PreviewOSS
    IBM · Open Source · via OpenRouter
    		82.4%$0.13
    Nova Micro
    Amazon · Proprietary
    		81.1%$0.17
    Codestral-22BOSS
    Mistral · Open Source · via Mistral AI
    		81.1%$1.20
    Llama 3.1 70B InstructOSS
    Meta · Open Source
    		80.5%$0.40
    Qwen2 7B InstructOSS
    Qwen · Open Source
    		79.9%$0.60
    Qwen2.5-Omni-7BOSS
    Qwen · Open Source
    		78.7%$0.50
    Claude 3 Haiku
    Anthropic · Proprietary
    		75.9%$1.50
    MiMo-V2.5-ProOSS
    Xiaomi · Open Source
    		75.6%$1.30
    Gemma 3n E4B Instructed
    Google · Proprietary
    		75%$60.00
    Showing 1–50 of 345 models

    Next step

    You found the model. Now ship the product.

    Auth, billing, and the API layer are already decided. Fork 8 finished AI apps, or have us build the first version with you.

    Build with the APIsHave us build it with you

    The index

    All Large Language Models

    345 models across 36 providers. Search or jump to a lab — every model page stays linked here.

    OpenAI52Anthropic22Google34Cursor2xAI17DeepSeek20MiniMax7Qwen55AI21 Labs2Amazon7Baidu2ByteDance4Cohere4IBM2Inception1inclusionAI1LG AI Research1Liquid AI2Meituan4Meta12Microsoft9Mistral26Moonshot AI11Nous Research1NVIDIA9OpenBMB1Poolside2Sakana AI1Sarvam AI2StepFun2Tencent1Thinking Machines1Unisound1Upstage1Xiaomi5Z AI21

    Baidu

    2 models
    ERNIE 5.0ERNIE 4.5

    Inception

    1 models
    Mercury 2

    inclusionAI

    1 models
    Ling-3.0-flash

    LG AI Research

    1 models
    K-EXAONE-236B-A23B

    Liquid AI

    2 models
    LFM2.5-VL-3BLFM2.5-2.6B

    Nous Research

    1 models
    Hermes 3 70B

    OpenBMB

    1 models
    MiniCPM-SALA

    Sakana AI

    1 models
    Sakana Namazu

    Sarvam AI

    2 models
    Sarvam-105BSarvam-30B

    Tencent

    1 models
    Hy3

    Thinking Machines

    1 models
    Inkling-Small

    Unisound

    1 models
    U2

    Upstage

    1 models
    Solar Pro 4

    Which model leads HumanEval right now?

    As of August 31, 2026, MiniCPM-SALA by OpenBMB is #1 for HumanEval at 95.1%. Ranked by the HumanEval score This board also tracks HumanEval. Next on the same board: Kimi K2 0905 and Claude 3.5 Sonnet. This humaneval leaderboard ranks models by HumanEval. Scores come from public evals. Prices are the live API rates in the table above.

    Sources: HumanEval (https://github.com/openai/human-eval); OpenAI API pricing (https://developers.openai.com/api/docs/pricing); Anthropic Claude API pricing (https://platform.claude.com/docs/en/about-claude/pricing); Gemini API pricing (https://ai.google.dev/gemini-api/docs/pricing)

    Top 8 for humaneval. Ranked by the HumanEval score Input and output are dollars per million tokens.
    RankModelHumanEvalInput /MOutput /M
    1MiniCPM-SALA95.1%$0.10$0.10
    2Kimi K2 090594.5%$0.60$2.50
    3Claude 3.5 Sonnet93.7%$3.00$15.00
    4GPT-593.4%$1.25$10.00
    5Kimi K2 Instruct93.3%$0.50$0.50
    6Qwen2.5-Coder 32B Instruct92.7%$0.09$0.09
    7o1-mini92.4%$3.00$12.00
    8Sarvam-30B92.1%$0.30$0.30

    Related LLM leaderboards

    LLM LeaderboardCodingMathAgenticReasoningWritingScienceFinanceLong ContextVisionOpen Source

    Individual benchmark pages

    Rank one eval at a time. All LLM benchmarks.

    HLESWE BenchHumanity's Last Exam (No Tools)Humanity's Last Exam (With Tools)Terminal-Bench 2.0Terminal-BenchTerminal-Bench 2.1Terminal-Bench 2Terminal-Bench 3.0ARC-AGI-1 Verified

    HumanEval FAQ

    Who ranks #1 on the HumanEval leaderboard?

    As of August 31, 2026, MiniCPM-SALA by OpenBMB ranks #1 on HumanEval at 95.1%. API pricing is $0.10/M input and $0.10/M output.

    What are the top models on HumanEval?

    The current HumanEval ranking as of August 31, 2026 is 1. MiniCPM-SALA at 95.1%; 2. Kimi K2 0905 at 94.5%; 3. Claude 3.5 Sonnet at 93.7%.

    Which humaneval model is the cheapest?

    Llama 3.1 8B Instruct is the cheapest scored model on this humaneval leaderboard at $0.03/M input and $0.03/M output ($0.06 blended). MiniCPM-SALA still leads HumanEval at 95.1%.

    Should I always pick the #1 HumanEval model?

    Not automatically. MiniCPM-SALA leads HumanEval, but a cheaper scored model can be the better production choice if the quality gap is small. Use the table to weigh HumanEval against input/output price, context window, and related evals.

    How often is the HumanEval leaderboard updated?

    Scores and API prices on this page are refreshed from published evals and provider rates. The snapshot is labeled August 31, 2026. Treat it as a current index, not a one-off blog post.

    What is HumanEval?

    Measures functional code generation — the model writes Python functions from docstrings and is evaluated on unit tests. This page ranks models that have published a HumanEval score, with live API token prices on the same row. Official methodology: HumanEval (https://github.com/openai/human-eval).

    Where is the HumanEval leaderboard?

    This page is the HumanEval leaderboard. Models are sorted by HumanEval, with input and output token prices on the same row so you can weigh score against cost. Official boards often omit price; that comparison is the point of this index.

    How is this HumanEval ranking different from the official board?

    The official HumanEval page owns the methodology. This page keeps the published HumanEval score next to live API $/M so you can pick a production SKU, not only a trophy number. Source: https://github.com/openai/human-eval