LiveBench Reasoning Average
General QA · 2024-06-12
LiveBench leaderboard aggregate for livebench reasoning average. LiveBench is refreshed over time to reduce contamination and covers reasoning, coding, mathematics, language, data analysis, instruction following, and agentic coding.
Top models (higher is better)
| Model | Score |
|---|---|
| Opus 4.8 | 89.7 |
| GPT-5.5 | 89.7 |
| Sonnet 5 | 88.7 |
| Opus 4.6 | 88.7 |
| GPT-5.4 | 88.1 |
| Claude Fable 5 | 87.7 |
| Opus 4.7 | 87.2 |
| Sonnet 4.6 | 84.8 |
| Gemini 3.1 Pro Preview | 84.0 |
| Qwen3.7-Max | 83.3 |
| GPT-5.2 | 83.2 |
| Kimi K2.7 Code | 82.8 |
| DeepSeek-V4-Pro | 82.7 |
| Gemini 3.5 Flash | 82.0 |
| GPT-5.4 Nano | 81.1 |
| Opus 4.5 | 80.1 |
| Kimi K2.6 | 79.4 |
| GLM-5.2 | 78.6 |
| GPT-5.2-Codex | 77.7 |
| Grok Build 0.1 | 76.4 |
| Qwen3.6 Plus (2026-04-02) | 75.8 |
| MiniMax M3 | 74.5 |
| GPT-5.4 Mini | 71.3 |
| Grok 4.3 | 70.8 |
| DeepSeek-V4-Flash | 70.6 |
| Qwen3.6 27B | 70.3 |