Anthropic
Claude 3 Opus
Top model of the Claude 3 family (Opus, Sonnet, Haiku). $15 in / $75 out per million tokens.
- Announced
- Access
- Public
Benchmark scores
| Test | Score | Best on the tracker | How it was measured |
|---|---|---|---|
| GPQA DiamondPhD-level science questionsPhD experts: 70%Saturated | 47% | Gemini 4 Argon 96% | Independentdefault settings (claude-3-opus-20240229)Run 27 Jan 2025 |
| OTIS Mock AIMEcompetition mathsSaturated | 4.7% | GPT-5.5 100% | Independentdefault settings (claude-3-opus-20240229)Run 25 Feb 2025 |
| METR time horizontask length an AI agent finishes half the timeNo longer updated | 4 min | Claude Mythos Preview 17 h | IndependentMETR Horizon v1.1 run 'claude_3_opus_inspect'; 95% CI 1.7–8.8 min |
Sources
- Introducing the next generation of ClaudeAnthropic · Official · anthropic.com
- Epoch AI Benchmarking Hub (model release dates and runs)Epoch AI · Dataset · 9 Nov 2026 · epoch.ai