Agentic benchmark
HLE w/ tools leaderboard
Humanity's Last Exam with tools. Every model the catalog carries a published HLE w/ tools value for, ranked by that value.
CategoryAgentic
MeasurePass@1
TasksExpert questions with tool use
DifficultyFrontier tool-augmented reasoning
Published byDeepSeek-V4 Technical Report
Tool-augmented Humanity's Last Exam scores reported in DeepSeek-V4 thinking-mode evaluations.
HLE w/ tools ranking
19 models with a published HLE w/ tools value, ordered by that value, highest first. Models the source has not scored on this benchmark are not listed — they are unmeasured, not last.
| Rank | Model | Provider | Pass@1 |
|---|---|---|---|
| 1 |  Claude Opus 5 | Anthropic | 64.7 |
| 2 |  DeepSeek V4.1 Flash | DeepSeek | 63.9 |
| 3 |  GLM-5.3 | Z.AI | 62.5 |
| 4 | DeepSeek V4 Pro 0813 | DeepSeek | 60.0 |
| 5 | Claude Sonnet 5 | Anthropic | 57.4 |
| 6 |  GPT-6 Astra | OpenAI | 57.2 |
| 7 |  Qwen3.8 Max | Alibaba | 56.2 |
| 8 | APApodex 1.1 | Apodex | 56.1 |
| 8 | OAOrnith-1.5-397B | Ornith AI | 56.1 |
| 10 |  Hy4 preview | Tencent | 55.4 |
| 11 | GLM-5.3-Flash | Z.AI | 55.3 |
| 12 | Qwen3.7 Max | Alibaba | 53.5 |
| 13 | DSdots3-note Preview | Dots Studio | 52.6 |
| 14 |  Agents-A1 | InternScience | 47.6 |
| 15 |  Step 3.7 Flash | StepFun | 47.2 |
| 16 | DeepSeek V4 Flash 0731 | DeepSeek | 45.1 |
| 17 |  Nemotron 3 Ultra | NVIDIA | 37.4 |
| 18 | OAOrnith-1.5-35B-A3B | Ornith AI | 33.4 |
| 19 | OAOrnith-1.5-9B | Ornith AI | 30.5 |
Evidence key: Observed
Rows are ordered by the value DeepSeek-V4 Technical Report published, highest first. The source does not state whether a higher value is the better result, so this page does not either: for a benchmark that measures a rate of failure, read the table from the bottom.