Claude Opus 4.7
Anthropic llmOpus upgrade emphasising reliability on long-running coding tasks and self-verification, trained with deliberately reduced cyber capability.
overall ⌄
72.6
10 of 125 ranked
price · $/M tokens
5 / 25
in / out, as of 2026-07-25
to run it yourself ⌄
API only
no weights to run
Benchmarks
What its maker published, and what anyone else measured. Every figure links the document it came from.
| benchmark | published | measured |
|---|---|---|
| GPQA | 94.2 | — |
| SWE-bench | 87.6 thinking | — |
| SWE-bench Pro | 64.3 thinking | — |
| HLE | 46.9 max | — |
| BrowseComp | 79.3 max | — |
Cheaper, and at least as good
On a 1,000-in, 1,000-out call these cost less and rank no lower — a hard comparison to argue with, and a narrow one. Value covers what it misses.
| Claude Sonnet 5.5 Anthropic | $0.0120 | overall 72.7 |
| GPT-5.2 OpenAI | $0.0158 | overall 73.6 |
| Claude Opus 5.5 Anthropic | $0.0240 | overall 75.0 |
| Claude Opus 4.8 Anthropic | $0.0300 | overall 76.4 |
| Claude Opus 5 Anthropic | $0.0300 | overall 78.3 |
Lineage
No recorded lineage — root or standalone model.
Metadata
| org | Anthropic |
| released | 2026-04-16 |
| license | proprietary |