← All benchmarks
Claude Opus 5.5 (Medium, Default Fallback)
Anthropic · Released 2026-09-22 · Independent benchmark scores from Artificial Analysis
Intelligence Index
51.2
Output speed
74.8
tokens/sec
Blended price
$8.00
per 1M tokens
Time to first token
15.07s
Model metadata
| Model ID | ea9e0da6-169f-43c6-92f5-730a10d6ff8d |
|---|---|
| Slug | claude-opus-5-5-medium |
| Release date | 2026-09-22 |
| Provider | Anthropic |
| Provider slug | anthropic |
| Output speed | 74.8 tok/s |
| Time to first token | 15.07s |
| Time to first answer | 15.07s |
| Blended price ($/1M) | $8.00 |
| Input price ($/1M) | $4.00 |
| Output price ($/1M) | $20.00 |
All evaluation scores (18)
| aime | — |
|---|---|
| aime 25 | — |
| artificial analysis coding index | — |
| artificial analysis intelligence index | 51.20 |
| artificial analysis math index | — |
| gpqa | — |
| hle | 54.7% |
| ifbench | — |
| lcr | 84.3% |
| livecodebench | — |
| math 500 | — |
| mmlu pro | — |
| scicode | 59.3% |
| tau banking | — |
| tau2 | — |
| terminalbench hard | — |
| terminalbench v2 1 | — |
| terminalbench v4 0 | 52.5% |
Source: artificialanalysis.ai/models/claude-opus-5-5-medium
Benchmark data from Artificial Analysis.