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
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 system | Score |
|---|---|---|
| 1 | Claude Fable 5.2· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 73.2% |
| 2 | Claude Fable 5.1· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 65% |
| 3 | Claude Opus 5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 64.7% |
| 4 | Claude Fable 5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 64.5% |
| 5 | Claude Opus 5.5Opus 5.5 System Card capability section · no tools · auto thinking · 1M total-token cap· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 64.4% |
| 6 | GLM 5.3· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 62.5% |
| 7 | Muse Spark 1.1· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 62.1% |
| 8 | Claude Fable 5.1no tools· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 60.9% |
| 9 | Claude Fable 5.1Opus 5.5 System Card capability section · no tools· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 60.9% |
| 10 | Claude Fable 5.1System Card capability summary · adaptive thinking · max effort · default sampling · 5 trials · context ≤1M unless noted · no tools· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 60.9% |
| 11 | DeepSeek V4 Pro 0813· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 60% |
| 12 | Muse Spark· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 58.4% |
| 13 | Claude Opus 4.8· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 57.9% |
| 14 | Claude Fable 5no tools· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 57.8% |
| 15 | Claude Fable 5System Card capability summary · adaptive thinking · max effort · default sampling · 5 trials · context ≤1M unless noted · no tools· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 57.8% |
| 16 | Claude Sonnet 5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 57.4% |
| 17 | GPT-5.5 Pro· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 57.2% |
| 18 | Claude Sonnet 5.5adaptive thinking · max effort · default sampling · 5-trial mean · context ≤1M unless noted · no tools · auto thinking · 980K task budget· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 56.9% |
| 19 | Claude Opus 5no tools· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 56.6% |
| 20 | Claude Opus 5Opus 5.5 System Card capability section · no tools· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 56.6% |
| 21 | Claude Opus 5System Card capability summary · adaptive thinking · max effort · default sampling · 5 trials · context ≤1M unless noted · no tools· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 56.6% |
| 22 | Claude Fable 5Opus 5 System Card capability summary · adaptive thinking · max effort unless noted · no tools· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 56.5% |
| 23 | Claude Opus 5DeepSeek 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. | 56.3% |
| 24 | Claude Opus 5Opus 5 System Card capability summary · adaptive thinking · max effort unless noted · no tools· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 56.3% |
| 25 | Kimi K3· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 56% |
| 26 | Doubao Seed 2.1 Pro· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 55.7% |
| 27 | GLM 5.3 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 55.3% |
| 28 | Claude Opus 4.7· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 54.7% |
| 29 | GLM 5.2· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 54.7% |
| 30 | Doubao Seed 2.1 Turbo· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 54.6% |
| 31 | Claude Fable 5Qwen表:无工具· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 53.3% |
| 32 | Claude Opus 4.6· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 53.1% |
| 33 | GLM 5.1· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 52.3% |
| 34 | GPT 5.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 52.2% |
| 35 | Muse Spark 1.1Meta Model API · xhigh · 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. | 52.2% |
| 36 | Gemini 3.1 Pro Preview· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 51.4% |
| 37 | Kimi K3.1 Pro· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 51.1% |
| 38 | Kimi K2 Thinking· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 2 sources
| 51% |
| 39 | Kimi K2.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 2 sources
| 50.2% |
| 40 | Fugu UltraFugu official headline evaluation · text-only· Reported by Sakana AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 50% |
| 41 | Claude Opus 4.8Fable/Mythos 5 System Card capability summary · adaptive thinking · max effort · 5 trials unless noted · no tools· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 49.8% |
| 42 | Claude Opus 4.8max · 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. | 49.8% |
| 43 | Claude Opus 4.8Opus 4.8 System Card capability summary · adaptive thinking · max effort · 5 trials unless noted · no tools· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 49.8% |
| 44 | Claude Opus 4.8Opus 5 System Card capability summary · adaptive thinking · max effort unless noted · no tools· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 49.8% |
| 45 | Claude Sonnet 4.6· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 49% |
| 46 | Qwen3.5 27B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 2 sources
| 48.5% |
| 47 | Qwen3.5 122B-A10B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 2 sources
| 47.5% |
| 48 | Qwen3.5 35B A3B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 2 sources
| 47.4% |
| 49 | GPT-5.6 SolQwen表:无工具· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 47.2% |
| 50 | Claude Opus 4.7Gemini 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. | 46.9% |
Showing 1–50 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 system | Score |
|---|---|
| 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)