HMMT FebBrowse 296

HMMT Feb

Can the model solve difficult problems that require more than factual recall?

Version not specifiedExact reported variant
Reasoning8 ranked models11 reported values2 reportsHigher is better

Top rankings

One row per model, using its best reported score across effort settings.

8
RankModelBest scoreBest reported setting
1Qwen3.5-397B-A17BQwen87.9%Reported configurationQwen report1 value · 1 reportApr 21, 2026 · source
2Claude Opus 4.5Anthropic85.3%Reported configurationQwen report1 value · 1 reportApr 21, 2026 · source
3Qwen3.5-27BQwen84.3%Reported configurationQwen report2 values · 2 reportsApr 21, 2026 · source
4Qwen3.6-27BQwen84.3%Reported configurationQwen report1 value · 1 reportApr 21, 2026 · source
5Qwen3.6-35B-A3BQwen83.6%Reported configurationQwen report2 values · 2 reportsApr 21, 2026 · source
6Gemma4-26B-A4BGoogle79.0%Reported configurationQwen report1 value · 1 reportApr 15, 2026 · source
7Qwen3.5-35B-A3BQwen78.7%Reported configurationQwen report1 value · 1 reportApr 15, 2026 · source
8Gemma4-31BGoogle77.2%Reported configurationQwen report2 values · 2 reportsApr 21, 2026 · source

Effort curve

Every sourced cost-linked effort value for this exact version. Lines connect complete sweeps only.

0
No cost-linked effort sweep for this version.
Definition and comparison boundarypublic methodology

Can the model solve difficult problems that require more than factual recall?

Multi-step reasoning on academic, mathematical, or abstract problems. Higher is better. The value is the percentage reported in this lab's table.

Harness, tools, prompts, attempt count, and benchmark version can materially change the result. Compare within one reporting context.

Method / source