DSBench-Hard †Browse 296

DSBench-Hard †

Can the model complete substantial programming work under this benchmark's agent setup?

Version not specifiedExact reported variant
Coding5 ranked models5 reported values1 reportsHigher is better

Top rankings

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

5
RankModelBest scoreBest reported setting
1Claude Opus 4.8Anthropic71.7%Reported configurationDeepSeek report1 value · 1 reportJul 31, 2026 · source
2DeepSeek V4 Flash 0731DeepSeek59.6%max effortDeepSeek report1 value · 1 reportJul 31, 2026 · source
3GLM-5.2Z.ai54.5%Reported configurationDeepSeek report1 value · 1 reportJul 31, 2026 · source
4DeepSeek V4 ProDeepSeek31.1%Reported configurationDeepSeek report1 value · 1 reportJul 31, 2026 · source
5DeepSeek-V4-FlashDeepSeek25.8%Reported configurationDeepSeek report1 value · 1 reportJul 31, 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 boundaryinternal methodology

Can the model complete substantial programming work under this benchmark's agent setup?

Software implementation, debugging, or repository work under the published evaluation protocol. Higher is better. The value is the percentage reported in this lab's table.

This is a publisher-defined internal evaluation. The task set or grading details are not fully public, so treat it as directional evidence.

† DeepSeek's internal difficult coding-agent test set; the source retains the dagger marker.

Method / source