Exact model identity
Claude 3.5 Sonnet (claude-3-5-sonnet-20240620; BLADE)
Anthropic · version status: reported
Similar model names are never merged automatically. Evaluation membership and numeric result rows only reference the exact ID
blade-claude-3-5-sonnet-20240620.Evaluation settings
| Benchmark | Work | Run / group | Scope |
|---|---|---|---|
| BLADE Decision-Discrimination MCQ | BLADE: Benchmarking Language Model Agents for Data-Driven Science | blade-creator-decision-mcq blade-arxiv-v3-mcq-zero-shot-temperature-0 | full · n=188 |
| BLADE End-to-End Analysis Generation | BLADE: Benchmarking Language Model Agents for Data-Driven Science | blade-creator-paper blade-arxiv-v3-generation-one-turn-40-temperature-08 | full · n=12 |
| BLADE End-to-End Analysis Generation | BLADE: Benchmarking Language Model Agents for Data-Driven Science | blade-creator-react blade-arxiv-v3-generation-react-20-temperature-08 | full · n=12 |
Numeric results
| Benchmark | Work | Run / group | Metric | Value |
|---|---|---|---|---|
| BLADE End-to-End Analysis Generation | BLADE: Benchmarking Language Model Agents for Data-Driven Science | blade-creator-paper blade-arxiv-v3-generation-one-turn-40-temperature-08 | Decision-type weighted F1-score | 43.9 percent |
| BLADE End-to-End Analysis Generation | BLADE: Benchmarking Language Model Agents for Data-Driven Science | blade-creator-react blade-arxiv-v3-generation-react-20-temperature-08 | Decision-type weighted F1-score | 43.1 percent |