AnotherWrapper
    Build with the APIs
    LLMImageVideoSpeechTranscription

    Math

    🏆 Leaderboard

    As of August 21, 2026, GPT-5.2 is #1 for math at 100%. Ranked by AIME 2025: American invitational math contest problems. Best AI for math, ranked by AIME 2025, MATH-500, and FrontierMath. Compare contest math vs research-level scores next to API price.

    Updated August 21, 2026282 models33 providers
    Kimi K2-Thinking-0905OSS
    Moonshot AI · Open Source
    		100%————————$2.47
    GPT-5.2 Pro
    OpenAI · Proprietary
    		100%————————$189.00
    GPT-5.2
    OpenAI · Proprietary
    		100%73.2%—94%40.3%15%——18.8%$15.75
    Gemini 3 Pro
    Google · Proprietary
    		100%91.9%23.4%93.9%37.6%20%96.4%—18.8%$14.00
    Claude Opus 4.6
    Anthropic · Proprietary
    		99.8%88.4%——40.7%50%——22.9%$30.00
    Gemini 3 Flash
    Google · Proprietary
    		99.7%89.4%—93.3%35.6%15%——4.2%$3.50
    LongCat-Flash-Thinking-2601OSS
    Meituan · Open Source
    		99.6%————————$1.50
    GPT-5.1 High
    OpenAI · Proprietary
    		99.6%————————$11.25
    Qwen3.8 MaxOSS
    Qwen · Open Source
    		99.4%92.7%———58%———$8.00
    Nemotron 3 Nano (30B A3B)OSS
    NVIDIA · Open Source
    		99.2%————————$0.30
    Grok 4.6
    xAI · Proprietary
    		99.2%93.2%———51%———$8.00
    GPT OSS 20B HighOSS
    OpenAI · Open Source
    		98.7%46.0%———————$0.60
    GPT-5.1 Medium
    OpenAI · Proprietary
    		98.4%————————$11.25
    Seed 2.0 Pro
    ByteDance · Proprietary
    		98.3%————————$3.50
    Grok 4.5
    xAI · Proprietary
    		97.8%93.4%———30%———$8.00
    Claude Fable 5
    Anthropic · Proprietary
    		97.8%85.9%———77%———$60.00
    Step-3.5-FlashOSS
    StepFun · Open Source
    		97.3%————————$0.50
    Gemini 3.7 Flash
    Google · Proprietary
    		97.2%94.8%———58%———$4.50
    GPT-5.1 Codex High
    OpenAI · Proprietary
    		96.7%————————$11.25
    DeepSeek-V4-Pro-MaxOSS
    DeepSeek · Open Source
    		96.7%89.7%———10%———$5.22
    Kimi K2.6OSS
    Moonshot AI · Open Source
    		96.1%90.8%——39.0%16%——14.6%$4.93
    Kimi K2.5OSS
    Moonshot AI · Open Source
    		96.1%————————$3.68
    GLM-4.7OSS
    Z AI · Open Source
    		95.7%83.3%84.9%88.2%2.4%6%——0%$2.80
    Qwen3.7 Max
    Qwen · Proprietary
    		95.6%90.9%———26%———$5.00
    Kimi K2.7 CodeOSS
    Moonshot AI · Open Source
    		95.6%87.9%———————$4.93
    GPT-5.4
    OpenAI · Proprietary
    		95.6%89.4%——47.6%56%——27.1%$17.50
    Gemini 3.1 Pro
    Google · Proprietary
    		95.6%94.4%——36.9%26%——16.7%$17.50
    GPT-5 High
    OpenAI · Proprietary
    		94.6%————————$11.25
    GPT-5
    OpenAI · Proprietary
    		94.6%86.2%84.7%92.8%26.3%18%96%—12.5%$11.25
    DeepSeek-V4-Flash-0731OSS
    DeepSeek · Open Source
    		94.4%91.0%———56%———$1.76
    MiMo-V2-FlashOSS
    Xiaomi · Open Source
    		94.1%59.3%———————$0.40
    GPT-5.1 Thinking
    OpenAI · Proprietary
    		94%———26.7%————$11.25
    GPT-5.1 Instant
    OpenAI · Proprietary
    		94%———26.7%————$11.25
    GPT-5.1
    OpenAI · Proprietary
    		94%66.7%—93.0%26.7%———4.2%$11.25
    GLM-4.6OSS
    Z AI · Open Source
    		93.9%74.5%—89.8%3.8%———2.1%$2.80
    Qwen3.7-Plus
    Qwen · Proprietary
    		93.3%87.9%———————$1.60
    Qwen3.6 Plus
    Qwen · Proprietary
    		93.3%88.4%——26.2%———8.3%$3.50
    Grok 4.3
    xAI · Proprietary
    		93.3%88.8%———11%———$3.75
    GLM-5.1OSS
    Z AI · Open Source
    		93.3%89.9%——33.5%22.2%——12.5%$5.80
    Claude Opus 5
    Anthropic · Proprietary
    		93.3%92.9%———78%———$30.00
    Grok-3
    xAI · Proprietary
    		93.3%84.6%—91.3%——89.8%93.3%—$18.00
    DeepSeek-V3.2OSS
    DeepSeek · Open Source · via OpenRouter
    		93.1%————————$0.57
    K-EXAONE-236B-A23B
    LG AI Research · Proprietary
    		92.8%————————$1.60
    o4-mini
    OpenAI · Proprietary
    		92.7%81.4%97.8%93.4%10.7%—94.2%93.4%2.1%$5.50
    GPT OSS 120BOSS
    OpenAI · Open Source
    		92.6%78.5%—92.0%——94.8%——$0.54
    GPT OSS 120B HighOSS
    OpenAI · Open Source
    		92.5%75.8%———————$0.60
    Qwen3-235B-A22B-Thinking-2507OSS
    Qwen · Open Source
    		92.3%80.0%——8.5%———0%$3.30
    Grok 4 Fast
    xAI · Proprietary
    		92%————————$0.70
    Grok-4.1 Fast Reasoning
    xAI · Proprietary
    		91.9%84.3%—89.5%—4%———$0.70
    Grok-4
    xAI · Proprietary
    		91.7%87.5%—————94%—$18.00
    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.

    Build with the APIsHave 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

    What is the best AI for math?

    As of August 21, 2026, GPT-5.2 by OpenAI is #1 for math at 100%. Ranked by AIME 2025: American invitational math contest problems. This board also tracks AIME 2025, GPQA Diamond, MATH, MGSM. Next on the same board: GPT-5.2 Pro and Gemini 3 Pro. Related leaders: Gemini 3.7 Flash on GPQA Diamond at 94.8%; GLM-4.5 on MATH at 98.2%. This math leaderboard ranks models by AIME 2025. Scores come from public evals. Prices are the live API rates in the table above.

    Sources: 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 math. Ranked by AIME 2025: American invitational math contest problems. Input and output are dollars per million tokens.
    RankModelAIME 2025Input /MOutput /M
    1GPT-5.2100%$1.75$14.00
    2GPT-5.2 Pro100%$21.00$168.00
    3Gemini 3 Pro100%$2.00$12.00
    4Kimi K2-Thinking-0905100%$0.47$2.00
    5Claude Opus 4.699.8%$5.00$25.00
    6Gemini 3 Flash99.7%$0.50$3.00
    7LongCat-Flash-Thinking-260199.6%$0.30$1.20
    8GPT-5.1 High99.6%$1.25$10.00

    AIME vs FrontierMath

    AIME is contest math. MATH-500 is textbook problems. FrontierMath is the set that is supposed to stay difficult. If every model is in the 90s on AIME, sort FrontierMath. That is the gap people actually mean by best AI for math in 2026.

    Related LLM leaderboards

    LLM LeaderboardCodingAgenticReasoningWritingScienceFinanceLong 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

    Math FAQ

    Who ranks #1 on the Math leaderboard?

    As of August 21, 2026, GPT-5.2 by OpenAI ranks #1 on AIME 2025 at 100%. API pricing is $1.75/M input and $14.00/M output.

    What is the best AI for math?

    As of August 21, 2026, GPT-5.2 by OpenAI is #1 for math at 100%. Ranked by AIME 2025: American invitational math contest problems. This board also tracks AIME 2025, GPQA Diamond, MATH, MGSM. Next on the same board: GPT-5.2 Pro and Gemini 3 Pro. Related leaders: Gemini 3.7 Flash on GPQA Diamond at 94.8%; GLM-4.5 on MATH at 98.2%.

    What are the top models on AIME 2025?

    The current AIME 2025 ranking as of August 21, 2026 is 1. GPT-5.2 at 100%; 2. GPT-5.2 Pro at 100%; 3. Gemini 3 Pro at 100%.

    Which math model is the cheapest?

    Llama 3.1 8B Instruct is the cheapest scored model on this math leaderboard at $0.03/M input and $0.03/M output ($0.06 blended). GPT-5.2 still leads AIME 2025 at 100%.

    Should I always pick the #1 AIME 2025 model?

    Not automatically. GPT-5.2 leads AIME 2025, but a cheaper scored model can be the better production choice if the quality gap is small. Use the table to weigh AIME 2025 against input/output price, context window, and related evals.

    How often is the Math leaderboard updated?

    Scores and API prices on this page are refreshed from published evals and provider rates. The snapshot is labeled August 21, 2026. Treat it as a current index, not a one-off blog post.

    What is the best LLM for math?

    Rank models by AIME 2025 for contest math, MATH-500 for textbook problems, and FrontierMath for research-level items. The strongest math model is not always the cheapest API.

    What does AIME 2025 measure?

    AIME 2025 is a set of American Invitational Mathematics Examination problems. It is a common public signal for competition-level mathematical reasoning.

    Is FrontierMath harder than AIME?

    Yes. FrontierMath is designed to stay difficult after models saturate AIME and MATH. Use it when you care about the extreme tail, not high-school contest scores.

    What is FrontierMath?

    FrontierMath is a research-level math eval built to stay hard after models saturate AIME and MATH-500. Use it for the extreme tail, not for high-school contest bragging.

    Is the best AI for homework the same as the best math model?

    People search best AI for homework. This page ranks contest and research math, not a cheat sheet. A high AIME score means contest reasoning. It does not mean the model should do someone's take-home.

    ChatGPT vs Claude for math: which is better?

    Rank AIME, MATH-500, and FrontierMath on this table. The lead moves. Token price still matters if you are sampling many solutions.