Skip to content

Benchmark

Humanity's Last Exam

Scores are percentages. Higher is better.

Models and systems measured
120
Scores recorded
230
Sources
3
IQ weight
1.67

Score distribution

34%
Mean
34.8%
Median
15.5%
Std dev
12.5%
10th percentile
55.7%
90th percentile
3.7%
Min
3.7%73.2%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.67 for chat models, measured over 113 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#Model or systemScore
101DeepSeek V4.1 Flashtext-only subset · pass@1 · reasoning_effort 100 · temp 1.0 · top_p .95· Reported by DeepSeek· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.39.1%
102ERNIE 5.0· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim39%
103DeepSeek V4 Flash 0731DeepSeek V4.1 launch chart · exact public value · setting not further specified · text-only subset· Reported by DeepSeek· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.37.8%
104Gemini 3 Pro PreviewGemini 3 Pro Model Card · results as of November 2025 · no tools· Reported by Google DeepMind· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.37.5%
105Nemotron 3 Ultra· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim37.4%
106Hy3 (Deep Infra)Hy3 official Benchmark Appendix · highest reasoning tier unless noted · text-only · no tools· Reported by Tencent Hunyuan· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.37%
107DeepSeek V4.1 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim36.8%
108DeepSeek V4.1 FlashDeepSeek V4.1 launch chart · exact public value · setting not further specified · full set· Reported by DeepSeek· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.36.8%
109DeepSeek V4.1 Flashfull set · pass@1 · reasoning_effort 100 · temp 1.0 · top_p .95· Reported by DeepSeek· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.36.8%
110GPT-5.2 Pro· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim36.6%
111GPT-5.2 Prono tools· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.36.6%
112GPT-5.2 Prono tools · xhigh· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.36.6%
113Qwen3.8 Omni FlashGPT-4o judge· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.36.5%
114Kimi K2.6· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim36.4%
115Kimi K2.6Kimi K2.6 official benchmark table · thinking · temp 1.0 · top_p 1.0 · 262,144 context · text-only subset · no tools · footnote result· Reported by Moonshot AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.36.4%
116Beam 1.0 501B A23B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim36%
117Qwen 3.8 Flash Next· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim35.9%
118Qwen 3.8 Flash NextQwen3.8-Flash-Next release/model card · GPT-4o judge· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.35.9%
119Qwen3.8 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim35.9%
120Qwen3.8 FlashGPT-4o judge· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.35.9%
121DeepSeek V4 FlashDeepSeek V4 official model card · Think Max· Reported by DeepSeek· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.34.8%
122Kimi K2.6Kimi K2.6 official benchmark table · thinking · temp 1.0 · top_p 1.0 · 262,144 context · full set · no tools · max generation 98,304· Reported by Moonshot AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.34.7%
123Qwen3.7 Plus· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim34.7%
124Qwen3.7 PlusGPT-4o judge· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.34.7%
125Qwen3.7 PlusQwen3.7-Plus release · recommended reasoning-effort=xhigh system prompt· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.34.7%
126Claude Sonnet 4.6no tools · Sonnet 5 card rerun· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.34.6%
127GPT 5.2· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim34.5%
128GPT 5.2no tools· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.34.5%
129GPT 5.2no tools · xhigh· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.34.5%
130MiMo V2.5 Pro· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim34%
131MiMo V2.5 ProXiaomi MiMo-V2.5-Pro official launch table · without tools· Reported by Xiaomi MiMo· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.34%
132DeepSeek V4 Flash 0731GPT-4o judge· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.33.8%
133Gemini 3 FlashGemini 3.5 Flash Model Card · May 2026 · full text + multimodal set· Reported by Google DeepMind· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.33.7%
134Gemini 3 Flashno tools · full text, text + multimodal set · pass@1· Reported by Google DeepMind· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.33.7%
135Claude Sonnet 4.6Gemini 3.5 Flash Model Card · May 2026 · full text + multimodal set· Reported by Google DeepMind· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.33.2%
136Claude Sonnet 4.6no tools · adaptive thinking · max effort· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.33.2%
137NVIDIA Nemotron 3.5 Super VL 120B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim31.8%
138Inkling Small· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim31.6%
139Kimi K2.5text subset · no tools· Reported by Moonshot AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.31.5%
140GLM 5.1GLM-5.1 official release table· Reported by Z.ai· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.31%
141Claude Opus 4.5reasoning only · without search/tools · decontaminated · Sonnet 4.5 grader· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.30.8%
142Qwen 3.8 27B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim30.8%
143Qwen 3.8 27BGPT-4o judge· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.30.8%
144Qwen 3.8 27BQwen3.8-27B model card · published setting· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.30.8%
145DeepSeek V3.2 Speciale· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim30.6%
146DeepSeek V3.2 SpecialeSpeciale high-compute reasoning · temperature 1.0 · 128K context · pass@1· Reported by DeepSeek· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.30.6%
147GLM 5· OpenEvals Leaderboard data· Reported by the model's developer. The source relaying it did not verify it.30.5%
148GLM 5GLM-5 official release table· Reported by Z.ai· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.30.5%
149Kimi K2.5full set · no tools· Reported by Moonshot AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.30.1%
150DeepSeek V4 FlashDeepSeek V4 official model card · Think High· Reported by DeepSeek· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.29.4%

Showing 101–150 of 229 results.

Not ranked

The source marks these results as descriptive, not capability scores, so we list them by name instead of ranking them.

Model or systemScore
Muse Spark (version unspecified)Meta Model API · highest effort · version unspecified · no tools · full 2,500 questions · o3-mini judge· Reported by Meta· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.42.8%

Data credits: Data by LLM Stats · BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links. · OpenEvals/leaderboard-data, MIT (licence)