Compare models
Comparing 2models. Drop the URL into a doc — it's permalinked.
| Field | anthropic/claude-opus-4-5 | deepseek/deepseek-reasoner |
|---|---|---|
| Provider | anthropic | deepseek |
| Model ID | claude-opus-4-5 | deepseek-reasoner |
| Context | 200K | 64K |
| Max output | 32K | 8K |
| Input / 1M | $15.00 | $0.55 |
| Output / 1M | $75.00 | $2.19 |
| Cached input / 1M | $1.50 | $0.14 |
| Avg cost / 1M | $45.00 | $1.37 |
| Speed | 65 t/s | 30 t/s |
| Quality index | 80.0 | 76.0 |
| MMLU | 91.4 | 90.8 |
| GPQA | 79.6 | 71.5 |
| HumanEval | 92.7 | 86.4 |
| MATH | 96.5 | 97.3 |
| SWE-bench | 72.5 | — |
| Arena Elo | — | — |
| Tools | ✓ | — |
| Vision | ✓ | — |
| Thinking | ✓ | ✓ |
| Streaming | ✓ | ✓ |
| JSON mode | — | — |
| Structured output | ✓ | — |
| Prompt cache | ✓ | — |
Same data, in your terminal: relay models compare opus-4.5 r1