Claude 3.5 Sonnet v2: benchmark results
Provider: Anthropic. Access: API.
Aggregate Rating 1684 ± 41, rank #425 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MedAgentBench | 69.67 | Overall SR (imported) | 100 |
| ZEROBench-Sub | 25.5 | Score (imported) | 80 |
| Reward Hacking Benchmark (RHB) | 0.6 | Exploit rate (%) (self-reported) | 79.2 |
| GSO-Bench | 15.7 | Opt@1 (%) | 51.7 |
| LLM Chess (Saplin) | -98.67 | Native anchored-opponent Elo (signed points) | 34.2 |
| ZeroBench | 3 | Score (%) | 29.5 |
Interactive version: theaggregate.ai/model?slug=claude-3-5-sonnet-v2 · How It Works · Data refreshed daily, snapshot 2026-11-08.