Model
Claude Code
Anthropic · closed weights
Coding
Coding| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| PolyCodeEval | 22.4 | pass_rate | N/A | Independent | medium | arxiv.org |
Safety & Alignment
Safety & Alignment| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| HarnessSecurity-Bench | 54.2 | attack_success_rate | harnesssecurity-bench | Independent | medium | arxiv.org |
Reproduced by the community
Loading community runs…