Updated 3 h ago · first seen 11 Sept 2026
bench_01M293SPERF0D8QG7TSCGE8GGF
- Metric
- average score · %
- Direction
- Higher is better
- Results
- 456
Leaderboard 456 current results
Select models with +, then open Compare.
| # | Model | Score | Config | Evaluated | Source | Actions |
|---|---|---|---|---|---|---|
| #101Qwen3.7 MaxQwen | 83.34% | release=2026-06-25 · subtasks=["theory_of_mind","zebra_puzzle","spatial","logic_with_navigation"] · livebench_model_id=qwen3.7-max | 25 Jun 2026 | livebench.aiT2 | History | |
| #102Claude 4.6 Opus Thinking High EffortAnthropic | 83.27% | release=2026-06-25 · subtasks=["connections","plot_unscrambling","typos"] · livebench_model_id=claude-opus-4-6-thinking-auto-high-effort | 25 Jun 2026 | livebench.aiT2 | History | |
| #103gpt-5.2-2025-12-11-high | 83.21% | release=2026-06-25 · subtasks=["theory_of_mind","zebra_puzzle","spatial","logic_with_navigation"] · livebench_model_id=gpt-5.2-2025-12-11-high | 25 Jun 2026 | livebench.aiT2 | History | |
| #104gpt-6-astraOpenAI | 82.97% | release=2026-06-25 · subtasks=["consecutive_events","tablejoin","tablereformat"] · livebench_model_id=gpt-6-astra-max | 25 Jun 2026 | livebench.aiT2 | History | |
| #105Claude Fable 5 xHigh EffortAnthropic | 82.97% | release=2026-06-25 · aggregation=mean of category averages; category = mean of its subtasks · livebench_model_id=claude-fable-5-max-effort | 25 Jun 2026 | livebench.aiT2 | History | |
| #106gpt-5.6-lunaOpenAI | 82.92% | release=2026-06-25 · subtasks=["code_generation","code_completion"] · livebench_model_id=gpt-5.6-luna-max | 25 Jun 2026 | livebench.aiT2 | History | |
| #107gpt-5.6-terraOpenAI | 82.89% | release=2026-06-25 · subtasks=["connections","plot_unscrambling","typos"] · livebench_model_id=gpt-5.6-terra-max | 25 Jun 2026 | livebench.aiT2 | History | |
| #108Kimi K2.7 CodeMoonshot AI | 82.81% | release=2026-06-25 · subtasks=["theory_of_mind","zebra_puzzle","spatial","logic_with_navigation"] · livebench_model_id=kimi-k2.7-code | 25 Jun 2026 | livebench.aiT2 | History | |
| #109Smaug Mini | 82.81% | release=2026-06-25 · subtasks=["theory_of_mind","zebra_puzzle","spatial","logic_with_navigation"] · livebench_model_id=smaug-mini | 25 Jun 2026 | livebench.aiT2 | History | |
| #110Grok 4.5xAI | 82.8% | release=2026-06-25 · subtasks=["connections","plot_unscrambling","typos"] · livebench_model_id=grok-4.5 | 25 Jun 2026 | livebench.aiT2 | History | |
| #111muse-spark-1-3-xhighMeta AI | 82.79% | release=2026-06-25 · subtasks=["connections","plot_unscrambling","typos"] · livebench_model_id=muse-spark-1.3-xhigh | 25 Jun 2026 | livebench.aiT2 | History | |
| #112deepseek-v4-proDeepSeek | 82.69% | release=2026-06-25 · subtasks=["theory_of_mind","zebra_puzzle","spatial","logic_with_navigation"] · livebench_model_id=deepseek-v4-pro | 25 Jun 2026 | livebench.aiT2 | History | |
| #113gpt-5.4OpenAI | 82.63% | release=2026-06-25 · subtasks=["connections","plot_unscrambling","typos"] · livebench_model_id=gpt-5.4-xhigh | 25 Jun 2026 | livebench.aiT2 | History | |
| #114Smaug Agentic | 82.47% | release=2026-06-25 · subtasks=["code_generation","code_completion"] · livebench_model_id=smaug-agentic | 25 Jun 2026 | livebench.aiT2 | History | |
| #115gpt-6-astraOpenAI | 82.16% | release=2026-06-25 · aggregation=mean of category averages; category = mean of its subtasks · livebench_model_id=gpt-6-astra-max | 25 Jun 2026 | livebench.aiT2 | History | |
| #116gpt-5.5OpenAI | 82.15% | release=2026-06-25 · subtasks=["code_generation","code_completion"] · livebench_model_id=gpt-5.5-xhigh | 25 Jun 2026 | livebench.aiT2 | History | |
| #117claude-opus-4-7-xhigh-effort | 82.09% | release=2026-06-25 · subtasks=["code_generation","code_completion"] · livebench_model_id=claude-opus-4-7-xhigh-effort | 25 Jun 2026 | livebench.aiT2 | History | |
| #118DeepSeek-V4-Pro-0813DeepSeek | 82.07% | release=2026-06-25 · subtasks=["connections","plot_unscrambling","typos"] · livebench_model_id=deepseek-v4-pro-0813 | 25 Jun 2026 | livebench.aiT2 | History | |
| #119gemini-3.5-flash-high | 82.01% | release=2026-06-25 · subtasks=["theory_of_mind","zebra_puzzle","spatial","logic_with_navigation"] · livebench_model_id=gemini-3.5-flash-high | 25 Jun 2026 | livebench.aiT2 | History | |
| #120Claude 4.8 Opus Thinking Max EffortAnthropic | 81.83% | release=2026-06-25 · subtasks=["code_generation","code_completion"] · livebench_model_id=claude-opus-4-8-max-effort | 25 Jun 2026 | livebench.aiT2 | History | |
| #121muse-spark-1-3-xhighMeta AI | 81.59% | release=2026-06-25 · aggregation=mean of category averages; category = mean of its subtasks · livebench_model_id=muse-spark-1.3-xhigh | 25 Jun 2026 | livebench.aiT2 | History | |
| #122gpt-5.5OpenAI | 81.58% | release=2026-06-25 · subtasks=["consecutive_events","tablejoin","tablereformat"] · livebench_model_id=gpt-5.5-xhigh | 25 Jun 2026 | livebench.aiT2 | History | |
| #123claude-opus-5-max-effort | 81.45% | release=2026-06-25 · subtasks=["code_generation","code_completion"] · livebench_model_id=claude-opus-5-max-effort | 25 Jun 2026 | livebench.aiT2 | History | |
| #124Kimi K3Moonshot AI | 81.45% | release=2026-06-25 · subtasks=["code_generation","code_completion"] · livebench_model_id=kimi-k3 | 25 Jun 2026 | livebench.aiT2 | History | |
| #125Gemini 3.8 FlashGoogle | 81.41% | release=2026-06-25 · subtasks=["paraphrase","simplify","story_generation","summarize"] · livebench_model_id=gemini-3.8-flash-high | 25 Jun 2026 | livebench.aiT2 | History | |
| #126claude-opus-4-5-20251101-thinking-64k-high-effort | 81.26% | release=2026-06-25 · subtasks=["connections","plot_unscrambling","typos"] · livebench_model_id=claude-opus-4-5-20251101-thinking-64k-high-effort | 25 Jun 2026 | livebench.aiT2 | History | |
| #127GLM 5.3 FlashZ.ai (Zhipu AI) | 81.25% | release=2026-06-25 · subtasks=["AMPS_Hard","integrals_with_game","math_comp","olympiad"] · livebench_model_id=glm-5.3-flash | 25 Jun 2026 | livebench.aiT2 | History | |
| #128DeepSeek-V4.1-FlashDeepSeek | 81.19% | release=2026-06-25 · subtasks=["connections","plot_unscrambling","typos"] · livebench_model_id=deepseek-v4.1-flash-max | 25 Jun 2026 | livebench.aiT2 | History | |
| #129DeepSeek-V4.1-FlashDeepSeek | 81.11% | release=2026-06-25 · aggregation=mean of category averages; category = mean of its subtasks · livebench_model_id=deepseek-v4.1-flash-max | 25 Jun 2026 | livebench.aiT2 | History | |
| #130gpt-5.4-nanoOpenAI | 81.1% | release=2026-06-25 · subtasks=["theory_of_mind","zebra_puzzle","spatial","logic_with_navigation"] · livebench_model_id=gpt-5.4-nano-xhigh | 25 Jun 2026 | livebench.aiT2 | History | |
| #131muse-spark-1-3-xhighMeta AI | 81.06% | release=2026-06-25 · subtasks=["code_generation","code_completion"] · livebench_model_id=muse-spark-1.3-xhigh | 25 Jun 2026 | livebench.aiT2 | History | |
| #132gpt-5-6-solOpenAI | 81.05% | release=2026-06-25 · aggregation=mean of category averages; category = mean of its subtasks · livebench_model_id=gpt-5.6-sol-max | 25 Jun 2026 | livebench.aiT2 | History | |
| #133Claude Sonnet 5 xHigh EffortAnthropic | 80.68% | release=2026-06-25 · subtasks=["code_generation","code_completion"] · livebench_model_id=claude-sonnet-5-xhigh-effort | 25 Jun 2026 | livebench.aiT2 | History | |
| #134Claude Fable 5 xHigh EffortAnthropic | 80.54% | release=2026-06-25 · subtasks=["consecutive_events","tablejoin","tablereformat"] · livebench_model_id=claude-fable-5-max-effort | 25 Jun 2026 | livebench.aiT2 | History | |
| #135DeepSeek V4 Flash Vision ExpDeepSeek | 80.36% | release=2026-06-25 · subtasks=["connections","plot_unscrambling","typos"] · livebench_model_id=deepseek-v4-flash-vision-exp | 25 Jun 2026 | livebench.aiT2 | History | |
| #136gpt-6-astraOpenAI | 80.36% | release=2026-06-25 · subtasks=["code_generation","code_completion"] · livebench_model_id=gpt-6-astra-max | 25 Jun 2026 | livebench.aiT2 | History | |
| #137Claude Fable 5.1 Max EffortAnthropic | 80.28% | release=2026-06-25 · subtasks=["consecutive_events","tablejoin","tablereformat"] · livebench_model_id=claude-fable-5-1-max-effort | 25 Jun 2026 | livebench.aiT2 | History | |
| #138gpt-5.5OpenAI | 80.19% | release=2026-06-25 · aggregation=mean of category averages; category = mean of its subtasks · livebench_model_id=gpt-5.5-xhigh | 25 Jun 2026 | livebench.aiT2 | History | |
| #139claude-opus-4-5-20251101-thinking-64k-high-effort | 80.09% | release=2026-06-25 · subtasks=["theory_of_mind","zebra_puzzle","spatial","logic_with_navigation"] · livebench_model_id=claude-opus-4-5-20251101-thinking-64k-high-effort | 25 Jun 2026 | livebench.aiT2 | History | |
| #140claude-opus-5-max-effort | 80.09% | release=2026-06-25 · aggregation=mean of category averages; category = mean of its subtasks · livebench_model_id=claude-opus-5-max-effort | 25 Jun 2026 | livebench.aiT2 | History | |
| #141DeepSeek-V4.1-FlashDeepSeek | 80.04% | release=2026-06-25 · subtasks=["code_generation","code_completion"] · livebench_model_id=deepseek-v4.1-flash-max | 25 Jun 2026 | livebench.aiT2 | History | |
| #142qwen3-8-27b-non-reasoningAlibaba Group | 80.03% | release=2026-06-25 · subtasks=["theory_of_mind","zebra_puzzle","spatial","logic_with_navigation"] · livebench_model_id=qwen3.8-27b | 25 Jun 2026 | livebench.aiT2 | History | |
| #143Gemini 3.7 FlashGoogle | 79.93% | release=2026-06-25 · subtasks=["paraphrase","simplify","story_generation","summarize"] · livebench_model_id=gemini-3.7-flash-high | 25 Jun 2026 | livebench.aiT2 | History | |
| #144Smaug Agentic | 79.9% | release=2026-06-25 · subtasks=["consecutive_events","tablejoin","tablereformat"] · livebench_model_id=smaug-agentic | 25 Jun 2026 | livebench.aiT2 | History | |
| #145Qwen3.6 27BQwen | 79.87% | release=2026-06-25 · subtasks=["AMPS_Hard","integrals_with_game","math_comp","olympiad"] · livebench_model_id=qwen3.6-27b | 25 Jun 2026 | livebench.aiT2 | History | |
| #146GLM 5.3Z.ai (Zhipu AI) | 79.86% | release=2026-06-25 · subtasks=["connections","plot_unscrambling","typos"] · livebench_model_id=glm-5.3 | 25 Jun 2026 | livebench.aiT2 | History | |
| #147gpt-5-6-solOpenAI | 79.84% | release=2026-06-25 · subtasks=["consecutive_events","tablejoin","tablereformat"] · livebench_model_id=gpt-5.6-sol-max | 25 Jun 2026 | livebench.aiT2 | History | |
| #148gpt-5.2-2025-12-11-high | 79.81% | release=2026-06-25 · subtasks=["connections","plot_unscrambling","typos"] · livebench_model_id=gpt-5.2-2025-12-11-high | 25 Jun 2026 | livebench.aiT2 | History | |
| #149Qwen3.7 MaxQwen | 79.74% | release=2026-06-25 · subtasks=["connections","plot_unscrambling","typos"] · livebench_model_id=qwen3.7-max | 25 Jun 2026 | livebench.aiT2 | History | |
| #150Qwen 3.8 MaxQwen | 79.69% | release=2026-06-25 · subtasks=["connections","plot_unscrambling","typos"] · livebench_model_id=qwen3.8-max | 25 Jun 2026 | livebench.aiT2 | History | |
| #151Claude 4.8 Opus Thinking Max EffortAnthropic | 79.66% | release=2026-06-25 · subtasks=["connections","plot_unscrambling","typos"] · livebench_model_id=claude-opus-4-8-max-effort | 25 Jun 2026 | livebench.aiT2 | History | |
| #152claude-opus-4-5-20251101-thinking-64k-high-effort | 79.65% | release=2026-06-25 · subtasks=["code_generation","code_completion"] · livebench_model_id=claude-opus-4-5-20251101-thinking-64k-high-effort | 25 Jun 2026 | livebench.aiT2 | History | |
| #153Z.ai GLM 5.2Z.ai (Zhipu AI) | 79.65% | release=2026-06-25 · subtasks=["code_generation","code_completion"] · livebench_model_id=glm-5.2 | 25 Jun 2026 | livebench.aiT2 | History | |
| #154deepseek-v4-flashDeepSeek | 79.65% | release=2026-06-25 · subtasks=["AMPS_Hard","integrals_with_game","math_comp","olympiad"] · livebench_model_id=deepseek-v4-flash | 25 Jun 2026 | livebench.aiT2 | History | |
| #155Kimi K2.7 CodeMoonshot AI | 79.6% | release=2026-06-25 · subtasks=["AMPS_Hard","integrals_with_game","math_comp","olympiad"] · livebench_model_id=kimi-k2.7-code | 25 Jun 2026 | livebench.aiT2 | History | |
| #156muse-spark-1-3-xhighMeta AI | 79.57% | release=2026-06-25 · subtasks=["consecutive_events","tablejoin","tablereformat"] · livebench_model_id=muse-spark-1.3-xhigh | 25 Jun 2026 | livebench.aiT2 | History | |
| #157Smaug Agentic | 79.51% | release=2026-06-25 · aggregation=mean of category averages; category = mean of its subtasks · livebench_model_id=smaug-agentic | 25 Jun 2026 | livebench.aiT2 | History | |
| #158DeepSeek V4 Flash Vision ExpDeepSeek | 79.48% | release=2026-06-25 · subtasks=["consecutive_events","tablejoin","tablereformat"] · livebench_model_id=deepseek-v4-flash-vision-exp | 25 Jun 2026 | livebench.aiT2 | History | |
| #159Kimi K2.6 ThinkingMoonshot AI | 79.38% | release=2026-06-25 · subtasks=["theory_of_mind","zebra_puzzle","spatial","logic_with_navigation"] · livebench_model_id=kimi-k2.6-thinking | 25 Jun 2026 | livebench.aiT2 | History | |
| #160DeepSeek V4 Flash (0731)DeepSeek | 79.33% | release=2026-06-25 · subtasks=["consecutive_events","tablejoin","tablereformat"] · livebench_model_id=deepseek-v4-flash-0731 | 25 Jun 2026 | livebench.aiT2 | History | |
| #161gpt-5.4OpenAI | 79.31% | release=2026-06-25 · subtasks=["consecutive_events","tablejoin","tablereformat"] · livebench_model_id=gpt-5.4-xhigh | 25 Jun 2026 | livebench.aiT2 | History | |
| #162gpt-5.6-terraOpenAI | 79.31% | release=2026-06-25 · subtasks=["consecutive_events","tablejoin","tablereformat"] · livebench_model_id=gpt-5.6-terra-max | 25 Jun 2026 | livebench.aiT2 | History | |
| #163claude-sonnet-4-6-thinking-auto-medium-effort | 79.27% | release=2026-06-25 · subtasks=["code_generation","code_completion"] · livebench_model_id=claude-sonnet-4-6-thinking-auto-medium-effort | 25 Jun 2026 | livebench.aiT2 | History | |
| #164DeepSeek-V4.1-FlashDeepSeek | 79.25% | release=2026-06-25 · subtasks=["consecutive_events","tablejoin","tablereformat"] · livebench_model_id=deepseek-v4.1-flash-max | 25 Jun 2026 | livebench.aiT2 | History | |
| #165DeepSeek-V4-Pro-0813DeepSeek | 79.24% | release=2026-06-25 · subtasks=["consecutive_events","tablejoin","tablereformat"] · livebench_model_id=deepseek-v4-pro-0813 | 25 Jun 2026 | livebench.aiT2 | History | |
| #166Smaug Flash | 79.21% | release=2026-06-25 · subtasks=["connections","plot_unscrambling","typos"] · livebench_model_id=smaug-flash | 25 Jun 2026 | livebench.aiT2 | History | |
| #167Kimi K3Moonshot AI | 79.19% | release=2026-06-25 · aggregation=mean of category averages; category = mean of its subtasks · livebench_model_id=kimi-k3 | 25 Jun 2026 | livebench.aiT2 | History | |
| #168DeepSeek V4 Flash (0731)DeepSeek | 79.18% | release=2026-06-25 · subtasks=["connections","plot_unscrambling","typos"] · livebench_model_id=deepseek-v4-flash-0731 | 25 Jun 2026 | livebench.aiT2 | History | |
| #169Smaug Flash | 79.12% | release=2026-06-25 · subtasks=["consecutive_events","tablejoin","tablereformat"] · livebench_model_id=smaug-flash | 25 Jun 2026 | livebench.aiT2 | History | |
| #170Gemini 3.1 Pro Preview HighGoogle | 79.1% | release=2026-06-25 · subtasks=["paraphrase","simplify","story_generation","summarize"] · livebench_model_id=gemini-3.1-pro-preview-high | 25 Jun 2026 | livebench.aiT2 | History | |
| #171GLM 5.3 FlashZ.ai (Zhipu AI) | 78.95% | release=2026-06-25 · subtasks=["code_generation","code_completion"] · livebench_model_id=glm-5.3-flash | 25 Jun 2026 | livebench.aiT2 | History | |
| #172GLM 5.3Z.ai (Zhipu AI) | 78.95% | release=2026-06-25 · subtasks=["code_generation","code_completion"] · livebench_model_id=glm-5.3 | 25 Jun 2026 | livebench.aiT2 | History | |
| #173Gemini 3.7 FlashGoogle | 78.89% | release=2026-06-25 · subtasks=["code_generation","code_completion"] · livebench_model_id=gemini-3.7-flash-high | 25 Jun 2026 | livebench.aiT2 | History | |
| #174Gemini 3.7 FlashGoogle | 78.83% | release=2026-06-25 · aggregation=mean of category averages; category = mean of its subtasks · livebench_model_id=gemini-3.7-flash-high | 25 Jun 2026 | livebench.aiT2 | History | |
| #175Smaug Mini | 78.81% | release=2026-06-25 · subtasks=["consecutive_events","tablejoin","tablereformat"] · livebench_model_id=smaug-mini | 25 Jun 2026 | livebench.aiT2 | History | |
| #176Kimi K3Moonshot AI | 78.73% | release=2026-06-25 · subtasks=["consecutive_events","tablejoin","tablereformat"] · livebench_model_id=kimi-k3 | 25 Jun 2026 | livebench.aiT2 | History | |
| #177Z.ai GLM 5.2Z.ai (Zhipu AI) | 78.63% | release=2026-06-25 · subtasks=["theory_of_mind","zebra_puzzle","spatial","logic_with_navigation"] · livebench_model_id=glm-5.2 | 25 Jun 2026 | livebench.aiT2 | History | |
| #178Muse Spark 1.2Meta AI | 78.58% | release=2026-06-25 · subtasks=["connections","plot_unscrambling","typos"] · livebench_model_id=muse-spark-1.2-xhigh | 25 Jun 2026 | livebench.aiT2 | History | |
| #179Kimi K2.6 ThinkingMoonshot AI | 78.57% | release=2026-06-25 · subtasks=["code_generation","code_completion"] · livebench_model_id=kimi-k2.6-thinking | 25 Jun 2026 | livebench.aiT2 | History | |
| #180Gemini 3.1 Pro Preview HighGoogle | 78.54% | release=2026-06-25 · subtasks=["consecutive_events","tablejoin","tablereformat"] · livebench_model_id=gemini-3.1-pro-preview-high | 25 Jun 2026 | livebench.aiT2 | History | |
| #181gpt-5.4-miniOpenAI | 78.46% | release=2026-06-25 · subtasks=["AMPS_Hard","integrals_with_game","math_comp","olympiad"] · livebench_model_id=gpt-5.4-mini-xhigh | 25 Jun 2026 | livebench.aiT2 | History | |
| #182Qwen 3.8 MaxQwen | 78.46% | release=2026-06-25 · aggregation=mean of category averages; category = mean of its subtasks · livebench_model_id=qwen3.8-max | 25 Jun 2026 | livebench.aiT2 | History | |
| #183Grok Build 0.1xAI | 78.43% | release=2026-06-25 · subtasks=["AMPS_Hard","integrals_with_game","math_comp","olympiad"] · livebench_model_id=grok-build-0.1 | 25 Jun 2026 | livebench.aiT2 | History | |
| #184Qwen 3.8 MaxQwen | 78.41% | release=2026-06-25 · subtasks=["consecutive_events","tablejoin","tablereformat"] · livebench_model_id=qwen3.8-max | 25 Jun 2026 | livebench.aiT2 | History | |
| #185Inkling xHigh Effort | 78.35% | release=2026-06-25 · subtasks=["theory_of_mind","zebra_puzzle","spatial","logic_with_navigation"] · livebench_model_id=inkling-xhigh | 25 Jun 2026 | livebench.aiT2 | History | |
| #186claude-opus-4-7-xhigh-effort | 78.26% | release=2026-06-25 · subtasks=["consecutive_events","tablejoin","tablereformat"] · livebench_model_id=claude-opus-4-7-xhigh-effort | 25 Jun 2026 | livebench.aiT2 | History | |
| #187gpt-5.6-terraOpenAI | 78.25% | release=2026-06-25 · subtasks=["code_generation","code_completion"] · livebench_model_id=gpt-5.6-terra-max | 25 Jun 2026 | livebench.aiT2 | History | |
| #188GPT-5.2-CodexOpenAI | 78.2% | release=2026-06-25 · subtasks=["consecutive_events","tablejoin","tablereformat"] · livebench_model_id=gpt-5.2-codex | 25 Jun 2026 | livebench.aiT2 | History | |
| #189Qwen3.6 PlusQwen | 78.18% | release=2026-06-25 · subtasks=["code_generation","code_completion"] · livebench_model_id=qwen3.6-plus | 25 Jun 2026 | livebench.aiT2 | History | |
| #190Claude 4.6 Opus Thinking High EffortAnthropic | 78.18% | release=2026-06-25 · subtasks=["code_generation","code_completion"] · livebench_model_id=claude-opus-4-6-thinking-auto-high-effort | 25 Jun 2026 | livebench.aiT2 | History | |
| #191gemini-3.5-flash-high | 78.18% | release=2026-06-25 · subtasks=["code_generation","code_completion"] · livebench_model_id=gemini-3.5-flash-high | 25 Jun 2026 | livebench.aiT2 | History | |
| #192gpt-5.2-2025-12-11-high | 78.16% | release=2026-06-25 · subtasks=["consecutive_events","tablejoin","tablereformat"] · livebench_model_id=gpt-5.2-2025-12-11-high | 25 Jun 2026 | livebench.aiT2 | History | |
| #193deepseek-v4-proDeepSeek | 78.13% | release=2026-06-25 · subtasks=["connections","plot_unscrambling","typos"] · livebench_model_id=deepseek-v4-pro | 25 Jun 2026 | livebench.aiT2 | History | |
| #194Grok 4.6xAI | 78.04% | release=2026-06-25 · aggregation=mean of category averages; category = mean of its subtasks · livebench_model_id=grok-4.6 | 25 Jun 2026 | livebench.aiT2 | History | |
| #195gpt-5.6-lunaOpenAI | 78.03% | release=2026-06-25 · subtasks=["consecutive_events","tablejoin","tablereformat"] · livebench_model_id=gpt-5.6-luna-max | 25 Jun 2026 | livebench.aiT2 | History | |
| #196muse-spark-1-3-xhighMeta AI | 78% | release=2026-06-25 · subtasks=["paraphrase","simplify","story_generation","summarize"] · livebench_model_id=muse-spark-1.3-xhigh | 25 Jun 2026 | livebench.aiT2 | History | |
| #197gpt-5.4OpenAI | 77.97% | release=2026-06-25 · aggregation=mean of category averages; category = mean of its subtasks · livebench_model_id=gpt-5.4-xhigh | 25 Jun 2026 | livebench.aiT2 | History | |
| #198Muse Spark 1.2Meta AI | 77.96% | release=2026-06-25 · aggregation=mean of category averages; category = mean of its subtasks · livebench_model_id=muse-spark-1.2-xhigh | 25 Jun 2026 | livebench.aiT2 | History | |
| #199claude-sonnet-4-6-thinking-auto-medium-effort | 77.95% | release=2026-06-25 · subtasks=["consecutive_events","tablejoin","tablereformat"] · livebench_model_id=claude-sonnet-4-6-thinking-auto-medium-effort | 25 Jun 2026 | livebench.aiT2 | History | |
| #200gpt-5.6-terraOpenAI | 77.94% | release=2026-06-25 · aggregation=mean of category averages; category = mean of its subtasks · livebench_model_id=gpt-5.6-terra-max | 25 Jun 2026 | livebench.aiT2 | History |
Scores are reported as published, with their evaluation configuration (harness, prompting, judge). The bar is relative to the best score on this page. Results with different configs are not directly comparable — see methodology.
The config filter matches a value inside each result's configuration (server-side, `config=` on the API). Chips are the values shared by several rows on the first page; per-model identifiers are not offered.
Definition
- Category
- general
Source:AI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2observed 3 h agomedium
- Task
- contamination-limited, monthly refreshed questions across 6 categories
Source:AI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2observed 3 h agomedium
- Metric
- average score · %
Source:AI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2observed 3 h agomedium
- Direction
- Higher is better
- Paper
- https://arxiv.org/abs/2406.19314
Source:AI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2observed 3 h agomedium
- Website
Source:AI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2observed 3 h agomedium
Each value shows its source, tier and observation time. Missing rows mean no source stated them. How results are recorded →
As of
Rewind the record: see this entity's attributes exactly as AI Atlas knew them on a given day.
Claim history
Categorycategory1
| Value | Valid from → to | Status | Source | Confidence | Extractor |
|---|---|---|---|---|---|
| general | → current | current | AI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2 | medium | curated |
Metricmetric1
| Value | Valid from → to | Status | Source | Confidence | Extractor |
|---|---|---|---|---|---|
| average score | → current | current | AI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2 | medium | curated |
Paperpaper1
| Value | Valid from → to | Status | Source | Confidence | Extractor |
|---|---|---|---|---|---|
| https://arxiv.org/abs/2406.19314 | → current | current | AI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2 | medium | curated |
Tasktask1
| Value | Valid from → to | Status | Source | Confidence | Extractor |
|---|---|---|---|---|---|
| contamination-limited, monthly refreshed questions across 6 categories | → current | current | AI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2 | medium | curated |
Unitunit1
| Value | Valid from → to | Status | Source | Confidence | Extractor |
|---|---|---|---|---|---|
| % | → current | current | AI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2 | medium | curated |
Websitewebsite1
| Value | Valid from → to | Status | Source | Confidence | Extractor |
|---|---|---|---|---|---|
| https://livebench.ai | → current | current | AI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2 | medium | curated |
Claims are temporal and append-only: a new observation closes the previous claim (valid_to) instead of overwriting it. Conflicting claims from different sources are kept side by side and flagged — never averaged. Methodology →
| Source | Document | Type | Tier | Last observed | Snapshots |
|---|---|---|---|---|---|
| LiveBench | livebench.ai/table_2026_06_25.csv | leaderboard | T2· Quality secondary | 57 min ago | 1 |
Tier 1 = official/primary, 2 = quality secondary, 3 = community, 4 = unverified. Every snapshot is archived; see all sources and the methodology.