| Accuracy45.9%Reported 45.9% no tools · Unit % Published Unconfirmed | HLE without tools / max- Settings
- Thinking auto; max effort; no tools; 980k-token task budget; no context compaction.
- Unknowns
- Per-effort trial count is not stated in the chart caption where applicable.
- Tools
- Tools forbidden
- Reasoning budget
- max
- Framework
- not recorded
- Protocol
- Humanity’s Last Exam; no tools; 980k-token task budget; no context compaction.
- Conditions
- Max effort; this is the no-tools main-table result. The card’s effort figure also labels the same 45.9% max point.
| View this score in rankings |
|---|
| Accuracy38.6%Reported 38.6% · Unit % Published Unconfirmed | HLE with tools / low- Settings
- Thinking auto; low effort; web search, web fetch, programmatic tool calling, and code execution; 980k-token task budget; no context compaction; Opus 4.6 grader.
- Unknowns
- Per-effort trial count is not stated in the chart caption where applicable.
- Tools
- Tools allowed
- Reasoning budget
- low
- Framework
- not recorded
- Protocol
- Humanity’s Last Exam; with web search, web fetch, programmatic tool calling, and code execution; 980k-token task budget.
- Conditions
- Low effort; no context compaction; Opus 4.6 grader; point read from the labeled figure.
| View this score in rankings |
|---|
| Accuracy45%Reported 45.0% · Unit % Published Unconfirmed | HLE with tools / medium- Settings
- Thinking auto; medium effort; web search, web fetch, programmatic tool calling, and code execution; 980k-token task budget; no context compaction; Opus 4.6 grader.
- Unknowns
- Per-effort trial count is not stated in the chart caption where applicable.
- Tools
- Tools allowed
- Reasoning budget
- medium
- Framework
- not recorded
- Protocol
- Humanity’s Last Exam; with web search, web fetch, programmatic tool calling, and code execution; 980k-token task budget.
- Conditions
- Medium effort; no context compaction; Opus 4.6 grader; point read from the labeled figure.
| View this score in rankings |
|---|
| Accuracy50.1%Reported 50.1% · Unit % Published Unconfirmed | HLE with tools / high- Settings
- Thinking auto; high effort; web search, web fetch, programmatic tool calling, and code execution; 980k-token task budget; no context compaction; Opus 4.6 grader.
- Unknowns
- Per-effort trial count is not stated in the chart caption where applicable.
- Tools
- Tools allowed
- Reasoning budget
- high
- Framework
- not recorded
- Protocol
- Humanity’s Last Exam; with web search, web fetch, programmatic tool calling, and code execution; 980k-token task budget.
- Conditions
- High effort; no context compaction; Opus 4.6 grader; point read from the labeled figure.
| View this score in rankings |
|---|
| Accuracy54.8%Reported 54.8% · Unit % Published Unconfirmed | HLE with tools / xhigh- Settings
- Thinking auto; xhigh effort; web search, web fetch, programmatic tool calling, and code execution; 980k-token task budget; no context compaction; Opus 4.6 grader.
- Unknowns
- Per-effort trial count is not stated in the chart caption where applicable.
- Tools
- Tools allowed
- Reasoning budget
- xhigh
- Framework
- not recorded
- Protocol
- Humanity’s Last Exam; with web search, web fetch, programmatic tool calling, and code execution; 980k-token task budget.
- Conditions
- Xhigh effort; no context compaction; Opus 4.6 grader; point read from the labeled figure.
| View this score in rankings |
|---|
| Accuracy57.4%Reported 57.4% · Unit % Published Unconfirmed | HLE with tools / max- Settings
- Thinking auto; max effort; web search, web fetch, programmatic tool calling, and code execution; 980k-token task budget; no context compaction; Opus 4.6 grader.
- Unknowns
- Per-effort trial count is not stated in the chart caption where applicable.
- Tools
- Tools allowed
- Reasoning budget
- max
- Framework
- not recorded
- Protocol
- Humanity’s Last Exam; with web search, web fetch, programmatic tool calling, and code execution; 980k-token task budget.
- Conditions
- Max effort; no context compaction; Opus 4.6 grader; point read from the labeled figure.
| View this score in rankings |
|---|
| Accuracy30.9%Reported 30.9% · Unit % Published Unconfirmed | HLE without tools / low- Settings
- Thinking auto; low effort; no tools; 980k-token task budget; no context compaction.
- Unknowns
- Per-effort trial count is not stated in the chart caption where applicable.
- Tools
- Tools forbidden
- Reasoning budget
- low
- Framework
- not recorded
- Protocol
- Humanity’s Last Exam; no tools; 980k-token task budget.
- Conditions
- Low effort; no context compaction; point read from the labeled figure.
| View this score in rankings |
|---|
| Accuracy35.5%Reported 35.5% · Unit % Published Unconfirmed | HLE without tools / medium- Settings
- Thinking auto; medium effort; no tools; 980k-token task budget; no context compaction.
- Unknowns
- Per-effort trial count is not stated in the chart caption where applicable.
- Tools
- Tools forbidden
- Reasoning budget
- medium
- Framework
- not recorded
- Protocol
- Humanity’s Last Exam; no tools; 980k-token task budget.
- Conditions
- Medium effort; no context compaction; point read from the labeled figure.
| View this score in rankings |
|---|
| Accuracy39.7%Reported 39.7% · Unit % Published Unconfirmed | HLE without tools / high- Settings
- Thinking auto; high effort; no tools; 980k-token task budget; no context compaction.
- Unknowns
- Per-effort trial count is not stated in the chart caption where applicable.
- Tools
- Tools forbidden
- Reasoning budget
- high
- Framework
- not recorded
- Protocol
- Humanity’s Last Exam; no tools; 980k-token task budget.
- Conditions
- High effort; no context compaction; point read from the labeled figure.
| View this score in rankings |
|---|
| Accuracy44.1%Reported 44.1% · Unit % Published Unconfirmed | HLE without tools / xhigh- Settings
- Thinking auto; xhigh effort; no tools; 980k-token task budget; no context compaction.
- Unknowns
- Per-effort trial count is not stated in the chart caption where applicable.
- Tools
- Tools forbidden
- Reasoning budget
- xhigh
- Framework
- not recorded
- Protocol
- Humanity’s Last Exam; no tools; 980k-token task budget.
- Conditions
- Xhigh effort; no context compaction; point read from the labeled figure.
| View this score in rankings |
|---|