HyperAI
Command Palette
Search for a command to run...
Humanity's Last Exam
数学、科学、人文科学にわたる2,500の専門家検証済み問題からなるフロンティアレベルのベンチマーク。最終的な閉形式の学術評価として設計されている。
- モデル
- 267 現在、 684 合計
- 中央値
- 12.0%
- 範囲
- 1.1%–74.3%
| ランク | 比較する | モデル | スコア |
|---|---|---|---|
| 151 |  Apriel-v1.6-15B-Thinker ServiceNow | 10.8% | |
| 152 | 10.6% | ||
| 153 |  Nemotron 3.5 Lightning NVIDIA | 10.6% | |
| 154 |  K2-V2 (High) Institute of Foundation Models | 10.5% | |
| 155 |  Solar Open 100B (Reasoning) Upstage | 10.4% | |
| 156 |  Magistral Medium 1.2 Mistral | 10.3% | |
| 157 |  Qwen3 Coder Next Alibaba | 10.1% | |
| 158 | K2 Think V2 Institute of Foundation Models | 10.1% | |
| 159 | Qwen3.5 4B (Reasoning) Alibaba | 9.9% | |
| 160 | Mistral Small 4 (Reasoning) Mistral | 9.9% | |
| 161 | 9.7% | ||
| 162 |  Ring-flash-2.0 InclusionAI | 9.6% | |
| 163 | 9.4% | ||
| 164 | Qwen3.5 9B (Non-reasoning) Alibaba | 9.4% | |
| 165 | Ling 3.0 Tiny InclusionAI | 9.3% | |
| 166 |  Nova 2.0 Lite (Medium) Amazon | 9.0% | |
| 167 |  MiniCPM5-2B OpenBMB | 8.9% | |
| 168 |  Hermes 4 - Llama-3.1 70B (Reasoning) Nous Research | 8.8% | |
| 169 |  Motif-2-12.7B-Reasoning Motif Technologies | 8.8% | |
| 170 | 8.6% | ||
| 171 |  Mi:dm K 2.5 Pro Korea Telecom | 8.1% | |
| 172 | Qwen3.5 4B (Non-reasoning) Alibaba | 8.0% | |
| 173 | Qwen3 Next 80B A3B Instruct Alibaba | 7.6% | |
| 174 | Qwen3.5 Omni Flash Alibaba | 7.6% | |
| 175 |  Sarvam 30B (High) Sarvam | 7.5% | |
| 176 | 7.5% | ||
| 177 | 7.4% | ||
| 178 | Nova 2.0 Omni (Medium) Amazon | 7.0% | |
| 179 |  LFM2.5-8B-A1B Liquid AI | 6.9% | |
| 180 |  Celeris-1 Celeris | 6.8% | |
| 181 | LFM2.5-1.2B-Instruct Liquid AI | 6.7% | |
| 182 | 6.6% | ||
| 183 | MiniCPM5-1B (Reasoning) OpenBMB | 6.5% | |
| 184 | 6.4% | ||
| 185 |  JT-35B-Flash China Mobile | 6.4% | |
| 186 | JT-MINI China Mobile | 6.4% | |
| 187 | Magistral Small 1.2 Mistral | 6.4% | |
| 188 | 6.3% | ||
| 189 |  Olmo 3.1 32B Think Allen Institute for AI | 6.3% | |
| 190 | LFM2.5-1.2B-Thinking Liquid AI | 6.2% | |
| 191 | LFM2.5-2.6B Liquid AI | 6.2% | |
| 192 |  Exaone 4.0 1.2B (Reasoning) LG AI Research | 6.0% | |
| 193 | Olmo 3 7B Think Allen Institute for AI | 6.0% | |
| 194 |  Tri-21B-Think Trillion Labs | 5.9% | |
| 195 |  gpt-oss-120b (Low) OpenAI | 5.9% | |
| 196 |  LongCat Flash Lite LongCat | 5.8% | |
| 197 | Olmo 3 7B Instruct Allen Institute for AI | 5.8% | |
| 198 | Exaone 4.0 1.2B (Non-reasoning) LG AI Research | 5.7% | |
| 199 | 5.5% | ||
| 200 | 5.5% |
