AnotherWrapper
    $249
    LLMImageVideoSpeechTranscription

    GSM8K

    🏆 Leaderboard

    As of August 20, 2026, MiMo-V2.5-Pro is #1 for GSM8K at 99.6%. Ranked by the GSM8K score 36 models in this index have a published GSM8K score. Methodology: GSM8K (https://github.com/openai/grade-school-math). GSM8K leaderboard: rank models by GSM8K next to live API token prices. Official methodology: GSM8K (https://github.com/openai/grade-school-math).

    Updated August 20, 2026282 models33 providers
    MiMo-V2.5-ProOSS
    Xiaomi · Open Source
    		99.6%$1.30
    Kimi K2 InstructOSS
    Moonshot AI · Open Source
    		97.3%$1.00
    o1
    OpenAI · Proprietary
    		97.1%$75.00
    Llama 3.1 405B InstructOSS
    Meta · Open Source
    		96.8%$1.78
    Claude 3.5 Sonnet
    Anthropic · Proprietary
    		96.4%$18.00
    Claude 3.5 Sonnet
    Anthropic · Proprietary
    		96.4%$18.00
    Gemma 3 27BOSS
    Google · Open Source
    		95.9%$0.30
    Qwen2.5 72B InstructOSS
    Qwen · Open Source
    		95.8%$0.75
    DeepSeek-V2.5OSS
    DeepSeek · Open Source
    		95.1%$0.42
    Claude 3 Opus
    Anthropic · Proprietary
    		95%$90.00
    Nova Pro
    Amazon · Proprietary
    		94.8%$4.00
    Nova Lite
    Amazon · Proprietary
    		94.5%$0.30
    Gemma 3 12BOSS
    Google · Open Source
    		94.4%$0.15
    Qwen3 235B A22BOSS
    Qwen · Open Source
    		94.4%$0.20
    Mistral Large 2OSS
    Mistral · Open Source · via Mistral AI
    		93%$8.00
    Nova Micro
    Amazon · Proprietary
    		92.3%$0.17
    Claude 3 Sonnet
    Anthropic · Proprietary
    		92.3%$18.00
    Qwen2.5 7B InstructOSS
    Qwen · Open Source
    		91.6%$0.60
    GPT-4o mini
    OpenAI · Proprietary
    		91.3%$0.75
    Qwen2.5-Coder 32B InstructOSS
    Qwen · Open Source
    		91.1%$0.18
    Gemini 1.5 Pro
    Google · Proprietary
    		90.8%$12.50
    GPT-4
    OpenAI · Proprietary
    		90.0%$90.00
    Gemma 3 4BOSS
    Google · Open Source
    		89.2%$0.06
    Claude 3 Haiku
    Anthropic · Proprietary
    		88.9%$1.50
    Phi 4 MiniOSS
    Microsoft · Open Source · via OpenRouter
    		88.6%$0.43
    Jamba 1.5 LargeOSS
    AI21 Labs · Open Source
    		87%$10.00
    Phi-3.5-mini-instructOSS
    Microsoft · Open Source
    		86.2%$0.20
    Gemini 1.5 Flash
    Google · Proprietary
    		86.2%$0.75
    Llama 3.1 8B InstructOSS
    Meta · Open Source
    		82.4%$0.06
    Granite 3.3 8B InstructOSS
    IBM · Open Source
    		80.9%$1.00
    Llama 3.2 3B InstructOSS
    Meta · Open Source
    		77.7%$0.03lowest
    Jamba 1.5 MiniOSS
    AI21 Labs · Open Source
    		75.8%$0.60
    Gemma 2 27BOSS
    Google · Open Source · via OpenRouter
    		74%$1.30
    Command R+OSS
    Cohere · Open Source
    		70.7%$1.25
    GPT-3.5 Turbo
    OpenAI · Proprietary
    		57.8%$2.00
    ERNIE 4.5
    Baidu · Proprietary
    		25.2%$4.40
    ChatGPT-4o Latest
    OpenAI · Proprietary
    		—$12.50
    Claude 3.5 Haiku
    Anthropic · Proprietary
    		—$4.80
    Claude 3.7 Sonnet
    Anthropic · Proprietary
    		—$18.00
    Claude Fable 5
    Anthropic · Proprietary
    		—$60.00
    Claude Haiku 4.5
    Anthropic · Proprietary
    		—$6.00
    Claude Mythos 5
    Anthropic · Proprietary
    		—$60.00
    Claude Mythos Preview
    Anthropic · Proprietary
    		—$60.00
    Claude Opus 4
    Anthropic · Proprietary
    		—$90.00
    Claude Opus 4.1
    Anthropic · Proprietary
    		—$90.00
    Claude Opus 4.5
    Anthropic · Proprietary
    		—$30.00
    Claude Opus 4.6
    Anthropic · Proprietary
    		—$30.00
    Claude Opus 4.7
    Anthropic · Proprietary
    		—$30.00
    Claude Opus 4.8
    Anthropic · Proprietary
    		—$30.00
    Claude Opus 5
    Anthropic · Proprietary
    		—$30.00
    Showing 1–50 of 282 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.

    Get the stack — $249Have us build it with you

    The index

    All Large Language Models

    282 models across 33 providers. Search or jump to a lab — every model page stays linked here.

    OpenAI51Anthropic22Google25Cursor2xAI15DeepSeek16MiniMax6Qwen38AI21 Labs2Amazon5Baidu1ByteDance2Cohere3IBM1Inception1inclusionAI1LG AI Research1Meituan4Meta12Microsoft5Mistral23Moonshot AI9Nous Research1NVIDIA5Poolside2Sakana AI1StepFun1Tencent1Thinking Machines1Unisound1Upstage1Xiaomi3Z AI20

    Baidu

    1 models
    ERNIE 4.5

    Inception

    1 models
    Mercury 2

    inclusionAI

    1 models
    Ling-3.0-flash

    LG AI Research

    1 models
    K-EXAONE-236B-A23B

    Nous Research

    1 models
    Hermes 3 70B

    Sakana AI

    1 models
    Sakana Namazu

    StepFun

    1 models
    Step-3.5-Flash

    Tencent

    1 models
    Hy3

    Thinking Machines

    1 models
    Inkling-Small

    Unisound

    1 models
    U2

    Upstage

    1 models
    Solar Pro 4

    Which model leads GSM8K right now?

    As of August 20, 2026, MiMo-V2.5-Pro by Xiaomi is #1 for GSM8K at 99.6%. Ranked by the GSM8K score This board also tracks GSM8K. Next on the same board: Kimi K2 Instruct and o1. This gsm8k leaderboard ranks models by GSM8K. Scores come from public evals. Prices are the live API rates in the table above.

    Sources: GSM8K (https://github.com/openai/grade-school-math); 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 gsm8k. Ranked by the GSM8K score Input and output are dollars per million tokens.
    RankModelGSM8KInput /MOutput /M
    1MiMo-V2.5-Pro99.6%$0.43$0.87
    2Kimi K2 Instruct97.3%$0.50$0.50
    3o197.1%$15.00$60.00
    4Llama 3.1 405B Instruct96.8%$0.89$0.89
    5Claude 3.5 Sonnet96.4%$3.00$15.00
    6Claude 3.5 Sonnet96.4%$3.00$15.00
    7Gemma 3 27B95.9%$0.10$0.20
    8Qwen2.5 72B Instruct95.8%$0.35$0.40

    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

    GSM8K FAQ

    Who ranks #1 on the GSM8K leaderboard?

    As of August 20, 2026, MiMo-V2.5-Pro by Xiaomi ranks #1 on GSM8K at 99.6%. API pricing is $0.43/M input and $0.87/M output.

    What are the top models on GSM8K?

    The current GSM8K ranking as of August 20, 2026 is 1. MiMo-V2.5-Pro at 99.6%; 2. Kimi K2 Instruct at 97.3%; 3. o1 at 97.1%.

    Which gsm8k model is the cheapest?

    Llama 3.2 3B Instruct is the cheapest scored model on this gsm8k leaderboard at $0.01/M input and $0.02/M output ($0.03 blended). MiMo-V2.5-Pro still leads GSM8K at 99.6%.

    Should I always pick the #1 GSM8K model?

    Not automatically. MiMo-V2.5-Pro leads GSM8K, but a cheaper scored model can be the better production choice if the quality gap is small. Use the table to weigh GSM8K against input/output price, context window, and related evals.

    How often is the GSM8K leaderboard updated?

    Scores and API prices on this page are refreshed from published evals and provider rates. The snapshot is labeled August 20, 2026. Treat it as a current index, not a one-off blog post.

    What is GSM8K?

    GSM8K is a public LLM eval (the GSM8K score). This page ranks models that have published a score, next to live API prices. Official methodology: GSM8K (https://github.com/openai/grade-school-math).

    Where is the GSM8K leaderboard?

    This page is the GSM8K leaderboard. Models are sorted by GSM8K, 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 GSM8K ranking different from the official board?

    The official GSM8K page owns the methodology. This page keeps the published GSM8K score next to live API $/M so you can pick a production SKU, not only a trophy number. Source: https://github.com/openai/grade-school-math