Exact model identity
Claude Code (Haiku 4.5)
Anthropic · version status: reported
Similar model names are never merged automatically. Evaluation membership and numeric result rows only reference the exact ID
claude-code-haiku-4-5.Evaluation settings
| Benchmark | Work | Run / group | Scope |
|---|---|---|---|
| CompBioBench | Agentic systems are adept at solving well-scoped, verifiable problems in computational biology | compbiobench-haiku-full compbiobench-v1-haiku-one-run-120m | full · n=100 |
| CompBioBench | Agentic systems are adept at solving well-scoped, verifiable problems in computational biology | compbiobench-haiku-hardest compbiobench-v1-hardest-haiku-one-run-120m | subset · n=17 |
Numeric results
| Benchmark | Work | Run / group | Metric | Value |
|---|---|---|---|---|
| CompBioBench | Agentic systems are adept at solving well-scoped, verifiable problems in computational biology | compbiobench-haiku-full compbiobench-v1-haiku-one-run-120m | Accuracy | 34 percent |
| CompBioBench | Agentic systems are adept at solving well-scoped, verifiable problems in computational biology | compbiobench-haiku-full compbiobench-v1-haiku-one-run-120m | Wall-clock time per question | 809.3 seconds |
| CompBioBench | Agentic systems are adept at solving well-scoped, verifiable problems in computational biology | compbiobench-haiku-full compbiobench-v1-haiku-one-run-120m | Cost per question | 0.3 USD |
| CompBioBench | Agentic systems are adept at solving well-scoped, verifiable problems in computational biology | compbiobench-haiku-hardest compbiobench-v1-hardest-haiku-one-run-120m | Accuracy — difficulty Levels 4–5 | 12 percent |