LiveCodeBench (Vals run)
Competitive programming problems collected after training cutoffs, easy to hard. Run by Vals AI.
| Category | Coding |
|---|---|
| Metric | Accuracy (%) |
| Source | Vals AI (evaluator) |
| License | None stated |
| Benchmark page | www.vals.ai |
| Status | Scored |
| Weight | 0.8 within Coding |
| Fit | Discrimination 0.59, noise 0.29 |
Results
| # | Model | Provider | Result | Date |
|---|---|---|---|---|
| 1 | Claude Fable 5.1 Max | Anthropic | 90.5% | 2026-09-01 |
| 2 | Claude Fable 5 Max | Anthropic | 89.8% | 2026-09-01 |
| 3 | Gemini 3.8 Flash High | 89.5% | 2026-09-01 | |
| 4 | Claude Opus 5 Max | Anthropic | 89.0% | 2026-09-01 |
| 5 | Gemini 3.7 Flash High | 88.7% | 2026-09-01 | |
| 6 | Gemini 3.1 Pro Preview High | 88.5% | 2026-09-01 | |
| 7 | Grok 4.6 High | xAI | 88.2% | 2026-09-01 |
| 8 | Gemini 3.6 Flash High | 88.1% | 2026-09-01 | |
| 9 | GPT-5.2 Codex | OpenAI | 88.0% | 2026-09-01 |
| 10 | Qwen3.8 Max | Alibaba | 87.9% | 2026-09-01 |
| 11 | Claude Opus 4.8 Max | Anthropic | 87.8% | 2026-09-01 |
| 12 | Gemini 3.5 Flash High | 87.6% | 2026-09-01 | |
| 13 | DeepSeek V4 Pro 0813 Max | DeepSeek | 87.5% | 2026-09-01 |
| 14 | DeepSeek V4 Pro Max | DeepSeek | 87.5% | 2026-09-01 |
| 15 | Grok 4.5 High | xAI | 87.4% | 2026-09-01 |
| 16 | GPT-5.3 Codex Extra High | OpenAI | 87.3% | 2026-09-01 |
| 17 | DeepSeek V4 Flash 0731 High | DeepSeek | 87.3% | 2026-09-01 |
| 18 | Kimi K3 Max | Moonshot AI | 87.2% | 2026-09-01 |
| 19 | Qwen3.7 Max | Alibaba | 87.1% | 2026-09-01 |
| 20 | Kimi K2.6 | Moonshot AI | 86.8% | 2026-09-01 |
| 21 | GPT-5 Mini High | OpenAI | 86.6% | 2026-09-01 |
| 22 | GPT-5.1 High | OpenAI | 86.5% | 2026-09-01 |
| 23 | Gemini 3 Pro Preview | 86.4% | 2026-09-01 | |
| 24 | Nemotron 3 Ultra 550B A55B | NVIDIA | 86.0% | 2026-09-01 |
| 25 | Qwen3.6 Plus | Alibaba | 86.0% | 2026-09-01 |
| 26 | Inkling Small | Thinking Machines | 85.9% | 2026-09-01 |
| 27 | GPT-5.6 Terra Extra High | OpenAI | 85.9% | 2026-09-01 |
| 28 | GPT-5 High | OpenAI | 85.9% | 2026-09-01 |
| 29 | Muse Spark 1.1 Extra High | Meta | 85.9% | 2026-09-01 |
| 30 | Gemini 3 Flash Preview High | 85.6% | 2026-09-01 | |
| 31 | GPT-5.1 Codex | OpenAI | 85.6% | 2026-09-01 |
| 32 | Inkling | Thinking Machines | 85.5% | 2026-09-01 |
| 33 | GPT-5.2 Extra High | OpenAI | 85.4% | 2026-09-01 |
| 34 | Qwen3.5 Plus Thinking | Alibaba | 85.3% | 2026-09-01 |
| 35 | GPT-5.5 Extra High | OpenAI | 85.3% | 2026-09-01 |
| 36 | Claude Opus 4.7 Max | Anthropic | 85.1% | 2026-09-01 |
| 37 | GPT-5-Codex High | OpenAI | 84.7% | 2026-09-01 |
| 38 | Claude Opus 4.6 Thinking | Anthropic | 84.7% | 2026-09-01 |
| 39 | Grok 4.3 High | xAI | 84.5% | 2026-09-01 |
| 40 | Grok 4.20 (Reasoning) | xAI | 84.3% | 2026-09-01 |
| 41 | GPT-5.4 Extra High | OpenAI | 84.1% | 2026-09-01 |
| 42 | GPT-5.4 nano High | OpenAI | 84.0% | 2026-09-01 |
| 43 | Qwen3.8 27B Extra High | Alibaba | 84.0% | 2026-09-01 |
| 44 | o3 High | OpenAI | 83.9% | 2026-09-01 |
| 45 | Kimi K2.5 Thinking | Moonshot AI | 83.9% | 2026-09-01 |
| 46 | Claude Opus 4.5 Thinking | Anthropic | 83.7% | 2026-09-01 |
| 47 | GPT-5.1 Codex Max High | OpenAI | 83.6% | 2026-09-01 |
| 48 | Qwen3.5 Flash | Alibaba | 83.3% | 2026-09-01 |
| 49 | Grok 4 | xAI | 83.2% | 2026-09-01 |
| 50 | GPT OSS 120B | OpenAI | 83.2% | 2026-09-01 |
| 51 | GPT-5.6 Sol Max | OpenAI | 82.6% | 2026-09-01 |
| 52 | Claude Sonnet 5 Max | Anthropic | 82.4% | 2026-09-01 |
| 53 | GLM-4.7 | Zhipu AI | 82.2% | 2026-09-01 |
| 54 | o4-mini High | OpenAI | 82.2% | 2026-09-01 |
| 55 | MiniMax-M3 | MiniMax | 82.2% | 2026-09-01 |
| 56 | Claude Sonnet 4.6 Max | Anthropic | 82.1% | 2026-09-01 |
| 57 | Kimi K2.7 Code | Moonshot AI | 82.1% | 2026-09-01 |
| 58 | GLM-5 Thinking | Zhipu AI | 81.9% | 2026-09-01 |
| 59 | MiniMax-M2.1 | MiniMax | 81.8% | 2026-09-01 |
| 60 | MiMo-V2.5 | Xiaomi | 81.5% | 2026-09-01 |
| 61 | GPT-5.4 mini Extra High | OpenAI | 81.5% | 2026-09-01 |
| 62 | GLM-5.1 | Zhipu AI | 81.4% | 2026-09-01 |
| 63 | MiMo-V2.5-Pro | Xiaomi | 81.4% | 2026-09-01 |
| 64 | GLM-4.6 | Zhipu AI | 81.0% | 2026-09-01 |
| 65 | DeepSeek V3.2 Thinking | DeepSeek | 80.7% | 2026-09-01 |
| 66 | Grok 4.1 Fast (Reasoning) | xAI | 80.6% | 2026-09-01 |
| 67 | GLM-5.3 Max | Zhipu AI | 80.5% | 2026-09-01 |
| 68 | GLM-5.3-Flash Max | Zhipu AI | 80.5% | 2026-09-01 |
| 69 | GPT OSS 20B | OpenAI | 80.4% | 2026-09-01 |
| 70 | Gemini 3.1 Flash Lite Preview | 80.1% | 2026-09-01 | |
| 71 | MiniMax-M2.7 | MiniMax | 79.9% | 2026-09-01 |
| 72 | MiniMax-M2.5 | MiniMax | 79.2% | 2026-09-01 |
| 73 | Gemini 2.5 Pro Preview 06-05 | 79.2% | 2026-09-01 | |
| 74 | Gemini 3.5 Flash Lite High | 79.0% | 2026-09-01 | |
| 75 | X-Ai/Grok-4-Fast-Reasoning | xAI | 79.0% | 2026-09-01 |
| 76 | Qwen3 Max | Alibaba | 78.2% | 2026-09-01 |
| 77 | Gemini 2.5 Flash (Sep 2025) Thinking | 76.2% | 2026-09-01 | |
| 78 | Gemini 2.5 Flash (Sep 2025) | 75.1% | 2026-09-01 | |
| 79 | Claude Opus 4.5 High | Anthropic | 75.0% | 2026-09-01 |
| 80 | Magistral Medium 1.2 | Mistral | 74.9% | 2026-09-01 |
| 81 | Claude Sonnet 4.5 Thinking | Anthropic | 73.0% | 2026-09-01 |
| 82 | Magistral Small 1.2 | Mistral | 72.1% | 2026-09-01 |
| 83 | o3-mini High | OpenAI | 71.5% | 2026-09-01 |
| 84 | Gemini 2.5 Flash-Lite Preview 09-2025 Thinking | 71.4% | 2026-09-01 | |
| 85 | Qwen3 235B-A22B | Alibaba | 70.6% | 2026-09-01 |
| 86 | Kimi K2 Instruct | Moonshot AI | 70.4% | 2026-09-01 |
| 87 | DeepSeek-R1 | DeepSeek | 70.2% | 2026-09-01 |
| 88 | GPT-5 Nano High | OpenAI | 70.2% | 2026-09-01 |
| 89 | Claude Opus 4 Thinking | Anthropic | 70.2% | 2026-09-01 |
| 90 | DeepSeek V3.2 | DeepSeek | 69.9% | 2026-09-01 |
| 91 | GLM-5.2 | Zhipu AI | 69.5% | 2026-09-01 |
| 92 | Laguna M.1 | Poolside | 68.1% | 2026-09-01 |
| 93 | Laguna XS.2 | Poolside | 67.8% | 2026-09-01 |
| 94 | Gemini 2.5 Flash-Lite Preview 09-2025 | 67.7% | 2026-09-01 | |
| 95 | GLM-4.5 | Zhipu AI | 67.4% | 2026-09-01 |
| 96 | Qwen3 Max Preview | Alibaba | 66.9% | 2026-09-01 |
| 97 | Claude Opus 4.1 Thinking | Anthropic | 66.5% | 2026-09-01 |
| 98 | DeepSeek V3 0324 | DeepSeek | 65.5% | 2026-09-01 |
| 99 | Claude Opus 4.1 | Anthropic | 64.6% | 2026-09-01 |
| 100 | Kimi K2 Thinking | Moonshot AI | 63.1% | 2026-09-01 |
| 101 | Claude Opus 4 | Anthropic | 62.6% | 2026-09-01 |
| 102 | Claude Sonnet 4 Thinking | Anthropic | 62.4% | 2026-09-01 |
| 103 | x-AI/Grok-Code-Fast 1 | xAI | 62.0% | 2026-09-01 |
| 104 | Claude Sonnet 3.7 Thinking | Anthropic | 60.4% | 2026-09-01 |
| 105 | Claude Sonnet 4 | Anthropic | 59.7% | 2026-09-01 |
| 106 | GPT-4.1 mini High | OpenAI | 58.2% | 2026-09-01 |
| 107 | Gemini 2.5 Flash (Apr 2025) | 56.9% | 2026-09-01 | |
| 108 | Claude Sonnet 3.7 | Anthropic | 56.7% | 2026-09-01 |
| 109 | Mistral Large 3 | Mistral | 55.3% | 2026-09-01 |
| 110 | GPT-4.1 High | OpenAI | 54.7% | 2026-09-01 |
| 111 | Grok 3 | xAI | 52.9% | 2026-09-01 |
| 112 | Devstral 2 | Mistral | 51.8% | 2026-09-01 |
| 113 | o1 High | OpenAI | 50.3% | 2026-09-01 |
| 114 | Claude Sonnet 3.5 v2 | Anthropic | 49.6% | 2026-09-01 |
| 115 | Gemini 2.5 Flash (Apr 2025) Thinking | 46.9% | 2026-09-01 | |
| 116 | X-Ai/Grok-4-Fast-Non-Reasoning | xAI | 46.1% | 2026-09-01 |
| 117 | Mistral Medium 3 | Mistral | 44.8% | 2026-09-01 |
| 118 | Gemini 2.0 Flash | 43.6% | 2026-09-01 | |
| 119 | GPT-4o (2024-11-20) High | OpenAI | 43.4% | 2026-09-01 |
| 120 | GPT-4.1 nano High | OpenAI | 42.7% | 2026-09-01 |
| 121 | Grok 4.1 Fast | xAI | 42.6% | 2026-09-01 |
| 122 | Claude Haiku 3.5 | Anthropic | 41.9% | 2026-09-01 |
| 123 | Gemini 1.5 Pro (Sept 2024) | 41.7% | 2026-09-01 | |
| 124 | Claude Haiku 4.5 Thinking | Anthropic | 41.2% | 2026-09-01 |
| 125 | Grok-2 (Dec 2024) | xAI | 38.7% | 2026-09-01 |
| 126 | Mistral Large 2.1 | Mistral | 37.1% | 2026-09-01 |
| 127 | Gemini 1.5 Flash (Sep 2024) | 36.9% | 2026-09-01 | |
| 128 | Command A | Cohere | 35.1% | 2026-09-01 |
| 129 | Mistral Small 3.1 | Mistral | 31.8% | 2026-09-01 |
| 130 | GPT-4o mini High | OpenAI | 26.4% | 2026-09-01 |
| 131 | mistral-small-2402 | Unknown | 15.8% | 2026-09-01 |