IFBench leaderboard
- Grok 4.3 leads
- 83.3%
- models measured
- 131
- lead over #2
- +0.5 pts
- median
- 63.1%
Leader on IFBench · read October 21
On October 21, 2026, 132 of our catalog's models had a IFBench result, and SpaceXAI Grok 4.3 led with 83.3%, ahead of MiniMax M3 at 82.9%.
Instructions with precise, checkable constraints the answer has to follow. Share followed. Run by Artificial Analysis on its own harness.
Role in the index
IFBench is shown as published and does not feed any of our indexes. Read the methodology
| # | Model | Result | Evidence | Coding index | Output $/Mtok |
|---|---|---|---|---|---|
| 1 | Grok 4.3xAI | 83.3% | 37.4 | $2.50 | |
| 2 | MiniMax M3MiniMax | 82.9% | 39.2 | $1.20 | |
| 3 | Nemotron 3 UltraNVIDIA | 81.4% | 30.0 | $2.40 | |
| 4 | Grok 4.20xAI | 81.2% | 33.3 | $2.50 | |
| 5 | Qwen3.7 MaxQwen | 80.5% | 54.1 | $4.425 | |
| 6 | MiMo-V2.5-ProXiaomi | 79.9% | 31.6 | $0.9152 | |
| 7 | Qwen3.5 397B A17BQwen | 78.8% | Not benchmarked | $2.34 | |
| 8 | Gemini 3 Flash PreviewGoogle | 78.0% | 43.7 | $3.00 | |
| 9 | Qwen3.7 PlusQwen | 78.0% | 33.2 | $1.28 | |
| 10 | GPT-5.2-CodexOpenAI | 77.6% | 45.4 | $14.00 | |
| 11 | Gemini 3.1 Flash LiteGoogle | 77.2% | Not benchmarked | $1.50 | |
| 12 | Gemini 3.1 Pro PreviewGoogle | 77.1% | 32.3 | $12.00 | |
| 13 | Gemini 3.5 FlashGoogle | 76.3% | 39.8 | $9.00 | |
| 14 | GLM 5.1Z.ai | 76.3% | 49.2 | $4.40 | |
| 15 | Kimi K2.6Moonshot AI | 76.0% | 52.1 | $4.00 | |
| 16 | GPT-5.4 NanoOpenAI | 75.9% | 44.8 | $1.25 | |
| 17 | GPT-5.5OpenAI | 75.9% | 71.3 | $30.00 | |
| 18 | MiniMax M2.7MiniMax | 75.7% | 36.2 | $1.20 | |
| 19 | Qwen3.5-122B-A10BQwen | 75.7% | Not benchmarked | $2.08 | |
| 20 | Gemma 4 31BGoogle | 75.6% | 39.0 | $0.40 | |
| 21 | Qwen3.5-27BQwen | 75.6% | Not benchmarked | $1.56 | |
| 22 | GPT-5 MiniOpenAI | 75.4% | 34.2 | $2.00 | |
| 23 | GPT-5.2OpenAI | 75.4% | 55.9 | $14.00 | |
| 24 | GPT-5.3-CodexOpenAI | 75.4% | 60.0 | $14.00 | |
| 25 | Qwen3.6 PlusQwen | 75.2% | 44.1 | $1.95 | |
| 26 | Command ACohere | 73.9% | Not benchmarked | $10.00 | |
| 27 | GPT-5.4OpenAI | 73.9% | 64.4 | $15.00 | |
| 28 | GLM 5.2Z.ai | 73.3% | 61.1 | $4.40 | |
| 29 | GPT-5.4 MiniOpenAI | 73.3% | 35.8 | $4.50 | |
| 30 | GLM 5 TurboZ.ai | 73.2% | Not benchmarked | $4.00 | |
| 31 | GPT-5OpenAI | 73.1% | 38.8 | $10.00 | |
| 32 | GPT-5.1OpenAI | 72.9% | 40.4 | $10.00 | |
| 33 | GPT-5.6 SolOpenAI | 72.7% | 73.5 | $20.00 | |
| 34 | Qwen3.5-35B-A3BQwen | 72.5% | Not benchmarked | $1.30 | |
| 35 | Gemma 4 26B A4BGoogle | 72.4% | Not benchmarked | $0.60 | |
| 36 | MiniMax M2MiniMax | 72.3% | Not benchmarked | $1.02 | |
| 37 | GLM 5Z.ai | 72.3% | 41.7 | $3.20 | |
| 38 | MiniMax M2.5MiniMax | 71.6% | 37.2 | $1.20 | |
| 39 | Nemotron 3 SuperNVIDIA | 71.5% | Not benchmarked | $0.45 | |
| 40 | o3OpenAI | 71.4% | Not benchmarked | $8.00 | |
| 41 | GPT-5.6 TerraOpenAI | 71.2% | 63.9 | $12.00 | |
| 42 | Solar Pro 3Upstage | 71.2% | Not benchmarked | $0.60 | |
| 43 | Nemotron 3 Nano 30B A3BNVIDIA | 71.1% | Not benchmarked | $0.22 | |
| 44 | o1OpenAI | 70.3% | Not benchmarked | $60.00 | |
| 45 | Kimi K2.5Moonshot AI | 70.2% | 38.6 | $2.85 | |
| 46 | GPT-5.1-CodexOpenAI | 70.0% | 15.3 | $10.00 | |
| 47 | MiniMax M2.1MiniMax | 69.9% | Not benchmarked | $1.20 | |
| 48 | Mercury 2Inception | 69.8% | Not benchmarked | $0.75 | |
| 49 | gpt-oss-120bOpenAI | 69.0% | Not benchmarked | $0.50 | |
| 50 | Mistral Medium 3.5Mistral AI | 68.8% | 25.5 | $7.50 | |
| 51 | o4 MiniOpenAI | 68.7% | Not benchmarked | $4.40 | |
| 52 | Kimi K2 ThinkingMoonshot AI | 68.1% | Not benchmarked | $2.50 | |
| 53 | GPT-5.1-Codex-MiniOpenAI | 67.9% | Not benchmarked | $2.00 | |
| 54 | GLM 4.7Z.ai | 67.9% | 55.1 | $2.20 | |
| 55 | GPT-5 NanoOpenAI | 67.6% | Not benchmarked | $0.40 | |
| 56 | Qwen3.6 27BQwen | 67.6% | 36.8 | $2.70 | |
| 57 | Step 3.7 FlashStepFun | 67.3% | Not benchmarked | $1.15 | |
| 58 | o3 MiniOpenAI | 67.1% | Not benchmarked | $4.40 | |
| 59 | MiMo-V2.5Xiaomi | 67.1% | 50.0 | $0.336 | |
| 60 | Qwen3.5-9BQwen | 66.7% | Not benchmarked | $0.15 | |
| 61 | Step 3.5 FlashStepFun | 66.5% | Not benchmarked | $0.30 | |
| 62 | gpt-oss-20bOpenAI | 65.1% | Not benchmarked | $0.15 | |
| 63 | Qwen3.6 35B A3BQwen | 64.4% | Not benchmarked | $1.00 | |
| 64 | Claude Fable 5Anthropic | 63.5% | 79.3 | $50.00 | |
| 65 | Kimi K2.7 CodeMoonshot AI | 63.1% | 53.4 | $4.00 | |
| 66 | Hy3 previewTencent | 63.1% | Not benchmarked | $0.60 | |
| 67 | Claude Opus 4.8Anthropic | 62.2% | 65.0 | $25.00 | |
| 68 | GLM 5V TurboZ.ai | 61.1% | Not benchmarked | $4.00 | |
| 69 | GLM 4.7 FlashZ.ai | 60.8% | Not benchmarked | $0.40 | |
| 70 | DeepSeek V3.2DeepSeek | 60.7% | 30.0 | $0.88 | |
| 71 | Qwen3 Next 80B A3B InstructQwen | 60.7% | Not benchmarked | $1.10 | |
| 72 | Claude Opus 4.7Anthropic | 58.6% | 67.5 | $25.00 | |
| 73 | Claude Opus 4.5Anthropic | 58.0% | 44.0 | $25.00 | |
| 74 | DeepSeek V3.1 TerminusDeepSeek | 57.0% | Not benchmarked | $1.0142 | |
| 75 | Claude Sonnet 4.6Anthropic | 56.6% | 57.3 | $15.00 | |
| 76 | Qwen3 VL 235B A22B InstructQwen | 56.5% | Not benchmarked | $1.90 | |
| 77 | Trinity Large ThinkingArcee AI | 56.3% | Not benchmarked | $0.80 | |
| 78 | DeepSeek V3.2 ExpDeepSeek | 54.1% | Not benchmarked | $0.41 | |
| 79 | Claude Opus 4.6Anthropic | 53.1% | 57.2 | $25.00 | |
| 80 | Qwen3 235B A22B Instruct 2507Qwen | 51.2% | Not benchmarked | $0.75 | |
| 81 | Qwen3 30B A3B Instruct 2507Qwen | 50.7% | Not benchmarked | $0.30 | |
| 82 | Gemini 2.5 FlashGoogle | 50.3% | Not benchmarked | $2.50 | |
| 83 | Gemini 2.5 Flash LiteGoogle | 49.9% | Not benchmarked | $0.40 | |
| 84 | Gemini 2.5 ProGoogle | 48.7% | 19.1 | $10.00 | |
| 85 | Mistral Small 4Mistral AI | 48.2% | Not benchmarked | $0.60 | |
| 86 | Llama 3.3 70B InstructMeta Llama | 47.1% | Not benchmarked | $0.72 | |
| 87 | Grok 3RetiredxAI | 46.9% | Not benchmarked | $15.00 | |
| 88 | Qwen3 VL 30B A3B InstructQwen | 45.1% | Not benchmarked | $0.70 | |
| 89 | GLM 4.5Z.ai | 44.1% | Not benchmarked | $2.20 | |
| 90 | GLM 4.6Z.ai | 43.4% | 8.8 | $2.20 | |
| 91 | Llama 4 MaverickMeta Llama | 43.0% | Not benchmarked | $1.00 | |
| 92 | GPT-4.1OpenAI | 43.0% | Not benchmarked | $8.00 | |
| 93 | Kimi K2 0905Moonshot AI | 41.7% | Not benchmarked | $2.50 | |
| 94 | Kimi K2 0711Moonshot AI | 41.5% | Not benchmarked | $2.30 | |
| 95 | Qwen3 30B A3BQwen | 41.5% | Not benchmarked | $0.50 | |
| 96 | DeepSeek V3 0324DeepSeek | 41.0% | Not benchmarked | $1.07 | |
| 97 | Qwen3 14BQwen | 40.5% | Not benchmarked | $0.23 | |
| 98 | Qwen3 Coder 480B A35BQwen | 40.5% | Not benchmarked | $1.50 | |
| 99 | Qwen3 VL 8B InstructQwen | 39.9% | Not benchmarked | $0.75 | |
| 100 | Mistral Medium 3.1Mistral AI | 39.8% | Not benchmarked | $2.00 | |
| 101 | R1 0528DeepSeek | 39.6% | Not benchmarked | $2.18 | |
| 102 | Llama 4 ScoutMeta Llama | 39.5% | Not benchmarked | $0.50 | |
| 103 | Mistral Medium 3Mistral AI | 39.3% | Not benchmarked | $2.00 | |
| 104 | GPT-4.1 MiniOpenAI | 38.3% | Not benchmarked | $1.60 | |
| 105 | Nova Pro 1.0Amazon | 38.1% | Not benchmarked | $3.20 | |
| 106 | GLM 4.5 AirZ.ai | 37.6% | Not benchmarked | $1.10 | |
| 107 | Qwen2.5 72B InstructQwen | 36.9% | Not benchmarked | $0.40 | |
| 108 | Gemma 3 12BGoogle | 36.7% | Not benchmarked | $0.1625 | |
| 109 | Qwen3 32BQwen | 36.3% | Not benchmarked | $0.40 | |
| 110 | Nova Premier 1.0Amazon | 36.2% | Not benchmarked | $12.50 | |
| 111 | GPT-4o (2024-08-06)OpenAI | 36.0% | Not benchmarked | $10.00 | |
| 112 | Qwen3 Coder NextQwen | 35.2% | Not benchmarked | $0.80 | |
| 113 | Llama 3.1 70B InstructMeta Llama | 34.4% | Not benchmarked | $0.56 | |
| 114 | GPT-4oOpenAI | 34.3% | Not benchmarked | $10.00 | |
| 115 | GLM 4.5VZ.ai | 34.2% | Not benchmarked | $1.80 | |
| 116 | Nova Lite 1.0Amazon | 34.1% | Not benchmarked | $0.24 | |
| 117 | GPT-4OpenAI | 33.2% | Not benchmarked | $60.00 | |
| 118 | Qwen3 Coder 30B A3B InstructQwen | 32.7% | Not benchmarked | $0.44 | |
| 119 | GPT-4.1 NanoOpenAI | 32.0% | Not benchmarked | $0.40 | |
| 120 | Gemma 3 27BGoogle | 31.8% | Not benchmarked | $0.30 | |
| 121 | Mistral Large 2407Mistral AI | 31.6% | Not benchmarked | $6.00 | |
| 122 | GPT-4o-miniOpenAI | 31.0% | Not benchmarked | $0.60 | |
| 123 | Reka Flash 3Reka | 30.4% | Not benchmarked | $0.20 | |
| 124 | GLM 4.6VZ.ai | 30.1% | Not benchmarked | $0.90 | |
| 125 | Nova Micro 1.0Amazon | 29.4% | Not benchmarked | $0.14 | |
| 126 | Llama 3.1 8B InstructMeta Llama | 28.6% | Not benchmarked | $0.08 | |
| 127 | Gemma 3 4BGoogle | 28.3% | Not benchmarked | $0.075 | |
| 128 | Mistral Small 3Mistral AI | 26.4% | Not benchmarked | $0.08 | |
| 129 | Llama 3.2 3B InstructMeta Llama | 26.2% | Not benchmarked | $0.3325 | |
| 130 | Granite 4.0 MicroIBM Granite | 24.8% | Not benchmarked | $0.112 | |
| 131 | Phi 4Microsoft | 23.5% | Not benchmarked | $0.14 | |
| 132 | Llama 3.2 1B InstructMeta Llama | 22.8% | Not benchmarked | $0.201 |
Only models in our catalog are listed, and each appears once with its best published run. 198 rows were read from the source on October 21, 2026, and rows that match no catalog model are left out.