Skip to content

Benchmark

SkillsBench · Copilot resolution-rate difference

Higher scores are better.

Models measured
4
Scores recorded
4
Sources
1
IQ weight
No weight published

Score distribution

-1
Mean
4
Median
5.8
Std dev
-8
10th percentile
5.4
90th percentile
-8
Min
-85.4
The band covers the middle 80% of scores. The line marks the median.

Leaderboard

#ModelScore
1GPT 5.5GitHub 2026-06-20 weekly suite · Copilot CLI vs Codex CLI with the same model · medium reasoning effort · five independent runs, best-scored run · 2h timeout · no web tools/tool search/MCP servers · adjusted task set, exact task count and version not stated· Reported by GitHub· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.5.4
2GPT 5.4GitHub 2026-06-20 weekly suite · Copilot CLI vs Codex CLI with the same model · medium reasoning effort · five independent runs, best-scored run · 2h timeout · no web tools/tool search/MCP servers · adjusted task set, exact task count and version not stated· Reported by GitHub· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.4
3Claude Sonnet 4.6GitHub 2026-06-20 weekly suite · Copilot CLI vs Claude Code with the same model · medium reasoning effort · five independent runs, best-scored run · 2h timeout · no web tools/tool search/MCP servers · adjusted task set, exact task count and version not stated· Reported by GitHub· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.-5.4
4Claude Opus 4.7GitHub 2026-06-20 weekly suite · Copilot CLI vs Claude Code with the same model · medium reasoning effort · five independent runs, best-scored run · 2h timeout · no web tools/tool search/MCP servers · adjusted task set, exact task count and version not stated· Reported by GitHub· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.-8

Data credits: BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links.