| Accuracy67.7%Reported 67.7% with tools · Unit % Published Unconfirmed | HLE 有工具- Settings
- Humanity's Last Exam; with tools; adaptive thinking/max effort per table note。
- Unknowns
- 具体题集和 trial count 未列。
- Tools
- Tools allowed
- Reasoning budget
- max
- Framework
- not recorded
- Protocol
- Humanity's Last Exam; with tools。
- Conditions
- 发布表未列题集版本。
| View this score in rankings |
|---|
| Accuracy52.9%Reported 52.9% · Unit % Published Unconfirmed | HLE no tools, low effort- Settings
- Reasoning-only; explicit low effort; adaptive thinking; 1M total-token cap; no context compaction; Claude Opus 4.6 grader.
- Unknowns
- The figure does not state the number of model trials.
- Tools
- Tools forbidden
- Reasoning budget
- low
- Framework
- not recorded
- Protocol
- HLE; no tools; low effort; adaptive thinking; 1M total-token cap; no context compaction; Claude Opus 4.6 grader.
- Conditions
- Exact task-set release and model-trial count are not stated.
| View this score in rankings |
|---|
| Accuracy59%Reported 59.0% · Unit % Published Unconfirmed | HLE no tools, medium effort- Settings
- Reasoning-only; explicit medium effort; adaptive thinking; 1M total-token cap; no context compaction; Claude Opus 4.6 grader.
- Unknowns
- The figure does not state the number of model trials.
- Tools
- Tools forbidden
- Reasoning budget
- medium
- Framework
- not recorded
- Protocol
- HLE; no tools; medium effort; adaptive thinking; 1M total-token cap; no context compaction; Claude Opus 4.6 grader.
- Conditions
- Exact task-set release and model-trial count are not stated.
| View this score in rankings |
|---|
| Accuracy59.5%Reported 59.5% · Unit % Published Unconfirmed | HLE no tools, high effort- Settings
- Reasoning-only; explicit high effort; adaptive thinking; 1M total-token cap; no context compaction; Claude Opus 4.6 grader.
- Unknowns
- The figure does not state the number of model trials.
- Tools
- Tools forbidden
- Reasoning budget
- high
- Framework
- not recorded
- Protocol
- HLE; no tools; high effort; adaptive thinking; 1M total-token cap; no context compaction; Claude Opus 4.6 grader.
- Conditions
- Exact task-set release and model-trial count are not stated.
| View this score in rankings |
|---|
| Accuracy62.8%Reported 62.8% · Unit % Published Unconfirmed | HLE no tools, xhigh effort- Settings
- Reasoning-only; explicit xhigh effort; adaptive thinking; 1M total-token cap; no context compaction; Claude Opus 4.6 grader.
- Unknowns
- The figure does not state the number of model trials.
- Tools
- Tools forbidden
- Reasoning budget
- xhigh
- Framework
- not recorded
- Protocol
- HLE; no tools; xhigh effort; adaptive thinking; 1M total-token cap; no context compaction; Claude Opus 4.6 grader.
- Conditions
- Exact task-set release and model-trial count are not stated.
| View this score in rankings |
|---|
| Accuracy64.4%Reported 64.4% · Unit % Published Unconfirmed | HLE no tools, max effort- Settings
- Reasoning-only; explicit max effort; adaptive thinking; 1M total-token cap; no context compaction; Claude Opus 4.6 grader.
- Unknowns
- The figure does not state the number of model trials.
- Tools
- Tools forbidden
- Reasoning budget
- max
- Framework
- not recorded
- Protocol
- HLE; no tools; max effort; adaptive thinking; 1M total-token cap; no context compaction; Claude Opus 4.6 grader.
- Conditions
- Exact task-set release and model-trial count are not stated. Keep separate from the with-tools score.
| View this score in rankings |
|---|
| Accuracy57.4%Reported 57.4% · Unit % Published Unconfirmed | HLE with tools, low effort- Settings
- Web search, web fetch, programmatic tool calling, and code execution; explicit low effort; adaptive thinking; 1M total-token cap; no context compaction; Claude Opus 4.6 grader.
- Unknowns
- The figure does not state the number of model trials.
- Tools
- Tools allowed
- Reasoning budget
- low
- Framework
- not recorded
- Protocol
- HLE; web search, web fetch, programmatic tools, and code execution; low effort; adaptive thinking; 1M total-token cap; no context compaction; Claude Opus 4.6 grader.
- Conditions
- Exact task-set release and model-trial count are not stated.
| View this score in rankings |
|---|
| Accuracy63%Reported 63.0% · Unit % Published Unconfirmed | HLE with tools, medium effort- Settings
- Web search, web fetch, programmatic tool calling, and code execution; explicit medium effort; adaptive thinking; 1M total-token cap; no context compaction; Claude Opus 4.6 grader.
- Unknowns
- The figure does not state the number of model trials.
- Tools
- Tools allowed
- Reasoning budget
- medium
- Framework
- not recorded
- Protocol
- HLE; web search, web fetch, programmatic tools, and code execution; medium effort; adaptive thinking; 1M total-token cap; no context compaction; Claude Opus 4.6 grader.
- Conditions
- Exact task-set release and model-trial count are not stated.
| View this score in rankings |
|---|
| Accuracy63.9%Reported 63.9% · Unit % Published Unconfirmed | HLE with tools, high effort- Settings
- Web search, web fetch, programmatic tool calling, and code execution; explicit high effort; adaptive thinking; 1M total-token cap; no context compaction; Claude Opus 4.6 grader.
- Unknowns
- The figure does not state the number of model trials.
- Tools
- Tools allowed
- Reasoning budget
- high
- Framework
- not recorded
- Protocol
- HLE; web search, web fetch, programmatic tools, and code execution; high effort; adaptive thinking; 1M total-token cap; no context compaction; Claude Opus 4.6 grader.
- Conditions
- Exact task-set release and model-trial count are not stated.
| View this score in rankings |
|---|
| Accuracy66.4%Reported 66.4% · Unit % Published Unconfirmed | HLE with tools, xhigh effort- Settings
- Web search, web fetch, programmatic tool calling, and code execution; explicit xhigh effort; adaptive thinking; 1M total-token cap; no context compaction; Claude Opus 4.6 grader.
- Unknowns
- The figure does not state the number of model trials.
- Tools
- Tools allowed
- Reasoning budget
- xhigh
- Framework
- not recorded
- Protocol
- HLE; web search, web fetch, programmatic tools, and code execution; xhigh effort; adaptive thinking; 1M total-token cap; no context compaction; Claude Opus 4.6 grader.
- Conditions
- Exact task-set release and model-trial count are not stated.
| View this score in rankings |
|---|