AnotherWrapper
    $249
    LLMImageVideoSpeechTranscription

    IFEval

    🏆 Leaderboard

    As of August 20, 2026, Qwen3.5-27B is #1 for IFEval at 95%. Ranked by IFEval, which checks whether the model followed the prompt (format, length, constraints). That is instruction-following, not 'sounds good.' 49 models in this index have a published IFEval score. Methodology: IFEval (https://github.com/google-research/google-research/tree/master/instruction_following_eval). IFEval leaderboard: rank models by IFEval next to live API token prices. Official methodology: IFEval (https://github.com/google-research/google-research/tree/master/instruction_following_eval).

    Updated August 20, 2026282 models33 providers
    Qwen3.5-27BOSS
    Qwen · Open Source
    		95%$2.70
    Qwen3.7-Plus
    Qwen · Proprietary
    		94.6%$1.60
    Qwen3.7 Max
    Qwen · Proprietary
    		94.3%$5.00
    Qwen3.6 Plus
    Qwen · Proprietary
    		94.3%$3.50
    o3-mini
    OpenAI · Proprietary
    		93.9%$5.50
    Qwen3.5-122B-A10BOSS
    Qwen · Open Source
    		93.4%$3.60
    Claude 3.7 Sonnet
    Anthropic · Proprietary
    		93.2%$18.00
    Qwen3.5-397B-A17BOSS
    Qwen · Open Source
    		92.6%$4.20
    Nova Pro
    Amazon · Proprietary
    		92.1%$4.00
    Llama 3.3 70B InstructOSS
    Meta · Open Source
    		92.1%$0.40
    Qwen3.5-35B-A3BOSS
    Qwen · Open Source
    		91.9%$2.25
    Qwen3.5-9BOSS
    Qwen · Open Source · via OpenRouter
    		91.5%$0.25
    Gemma 3 27BOSS
    Google · Open Source
    		90.4%$0.30
    Nemotron Nano 9B v2OSS
    NVIDIA · Open Source · via OpenRouter
    		90.3%—
    Gemma 3 4BOSS
    Google · Open Source
    		90.2%$0.06
    Kimi K2-Instruct-0905OSS
    Moonshot AI · Open Source · via OpenRouter
    		89.8%$3.10
    Kimi K2 InstructOSS
    Moonshot AI · Open Source
    		89.8%$1.00
    Nova Lite
    Amazon · Proprietary
    		89.7%$0.30
    LongCat-Flash-ChatOSS
    Meituan · Open Source
    		89.7%$1.50
    Qwen3-Next-80B-A3B-ThinkingOSS
    Qwen · Open Source
    		88.9%$1.65
    Gemma 3 12BOSS
    Google · Open Source
    		88.9%$0.15
    Qwen3-235B-A22B-Instruct-2507OSS
    Qwen · Open Source
    		88.7%$0.95
    Llama 3.1 405B InstructOSS
    Meta · Open Source
    		88.6%$1.78
    Qwen3 VL 235B A22B ThinkingOSS
    Qwen · Open Source
    		88.2%$3.94
    Qwen3-235B-A22B-Thinking-2507OSS
    Qwen · Open Source
    		87.8%$3.30
    Qwen3 VL 235B A22B InstructOSS
    Qwen · Open Source
    		87.8%$1.79
    Qwen3-Next-80B-A3B-InstructOSS
    Qwen · Open Source
    		87.6%$1.65
    Llama 3.1 70B InstructOSS
    Meta · Open Source
    		87.5%$0.40
    GPT-4.1
    OpenAI · Proprietary
    		87.4%$10.00
    Nova Micro
    Amazon · Proprietary
    		87.2%$0.17
    DeepSeek-V3OSS
    DeepSeek · Open Source
    		86.1%$1.37
    Qwen3 VL 30B A3B InstructOSS
    Qwen · Open Source
    		85.8%$0.90
    Qwen3 VL 32B InstructOSS
    Qwen · Open Source · via OpenRouter
    		84.7%$0.52
    Qwen2.5 72B InstructOSS
    Qwen · Open Source
    		84.1%$0.75
    GPT-4.1 mini
    OpenAI · Proprietary
    		84.1%$2.00
    Qwen3 VL 8B InstructOSS
    Qwen · Open Source
    		83.7%$0.58
    Qwen3 VL 8B ThinkingOSS
    Qwen · Open Source
    		83.2%$2.27
    Mistral Small 3 24B InstructOSS
    Mistral · Open Source · via Mistral AI
    		82.9%$0.21
    Qwen3 VL 4B ThinkingOSS
    Qwen · Open Source
    		82.6%$1.10
    Qwen3 VL 4B InstructOSS
    Qwen · Open Source
    		82.3%$0.70
    Qwen3 VL 30B A3B ThinkingOSS
    Qwen · Open Source
    		81.7%$1.20
    GPT-4o
    OpenAI · Proprietary
    		81%$12.50
    Llama 3.1 8B InstructOSS
    Meta · Open Source
    		80.4%$0.06
    Llama 3.2 3B InstructOSS
    Meta · Open Source
    		77.4%$0.03lowest
    Granite 3.3 8B InstructOSS
    IBM · Open Source
    		74.8%$1.00
    GPT-4.1 nano
    OpenAI · Proprietary
    		74.5%$0.50
    Qwen2.5 7B InstructOSS
    Qwen · Open Source
    		71.2%$0.60
    Phi 4OSS
    Microsoft · Open Source
    		63%$0.21
    Pixtral-12BOSS
    Mistral · Open Source · via Mistral AI
    		61.3%$0.30
    ChatGPT-4o Latest
    OpenAI · Proprietary
    		—$12.50
    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 IFEval right now?

    As of August 20, 2026, Qwen3.5-27B by Qwen is #1 for IFEval at 95%. Ranked by IFEval, which checks whether the model followed the prompt (format, length, constraints). That is instruction-following, not 'sounds good.' This board also tracks IFEval. Next on the same board: Qwen3.7-Plus and Qwen3.7 Max. This ifeval leaderboard ranks models by IFEval. Scores come from public evals. Prices are the live API rates in the table above.

    Sources: IFEval (https://github.com/google-research/google-research/tree/master/instruction_following_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 ifeval. Ranked by IFEval, which checks whether the model followed the prompt (format, length, constraints). That is instruction-following, not 'sounds good.' Input and output are dollars per million tokens.
    RankModelIFEvalInput /MOutput /M
    1Qwen3.5-27B95%$0.30$2.40
    2Qwen3.7-Plus94.6%$0.32$1.28
    3Qwen3.7 Max94.3%$1.25$3.75
    4Qwen3.6 Plus94.3%$0.50$3.00
    5o3-mini93.9%$1.10$4.40
    6Qwen3.5-122B-A10B93.4%$0.40$3.20
    7Claude 3.7 Sonnet93.2%$3.00$15.00
    8Qwen3.5-397B-A17B92.6%$0.60$3.60

    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

    IFEval FAQ

    Who ranks #1 on the IFEval leaderboard?

    As of August 20, 2026, Qwen3.5-27B by Qwen ranks #1 on IFEval at 95%. API pricing is $0.30/M input and $2.40/M output.

    What are the top models on IFEval?

    The current IFEval ranking as of August 20, 2026 is 1. Qwen3.5-27B at 95%; 2. Qwen3.7-Plus at 94.6%; 3. Qwen3.7 Max at 94.3%.

    Which ifeval model is the cheapest?

    Llama 3.2 3B Instruct is the cheapest scored model on this ifeval leaderboard at $0.01/M input and $0.02/M output ($0.03 blended). Qwen3.5-27B still leads IFEval at 95%.

    Should I always pick the #1 IFEval model?

    Not automatically. Qwen3.5-27B leads IFEval, but a cheaper scored model can be the better production choice if the quality gap is small. Use the table to weigh IFEval against input/output price, context window, and related evals.

    How often is the IFEval 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 IFEval?

    IFEval is a public LLM eval (IFEval, which checks whether the model followed the prompt (format, length, constraints). That is instruction-following, not 'sounds good.'). This page ranks models that have published a score, next to live API prices. Official methodology: IFEval (https://github.com/google-research/google-research/tree/master/instruction_following_eval).

    Where is the IFEval leaderboard?

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

    The official IFEval page owns the methodology. This page keeps the published IFEval score next to live API $/M so you can pick a production SKU, not only a trophy number. Source: https://github.com/google-research/google-research/tree/master/instruction_following_eval