claude-opus-5 · Released 24 July 2026
| Provider | Anthropic |
|---|---|
| Family | Claude 5 |
| Modality | multimodal |
| Context | — |
| Max output | — |
| Reasoning | Yes |
| Tool calling | Yes |
| Structured output | Yes |
| Open weights | No / not recorded |
| Licence | — |
No verified Cybatar-run benchmark result has been published for this model yet.
Versioned observations tied to the exact deployment and source date.
| Effective | Deployment | Mode | Input / 1M | Cached / 1M | Output / 1M | Source |
|---|---|---|---|---|---|---|
| 24 Jul 2026 | Anthropic | standard | USD 5 | — | USD 25 | Source ↗ |
Published Opus 5 API pricing.
Evidence ↗Official-source release record for the exact Claude Opus 5 model version.
Evidence ↗Opus 5 launches as Anthropic’s high-capability model for long-running agents, coding and knowledge work.
Evidence ↗Reported by independent evaluators or the model provider. These are contextual research observations, not Cybatar-run scores.
| Benchmark | Score | Configuration / harness | Provenance | Source |
|---|---|---|---|---|
| Artificial Analysis Intelligence Index v4.1.1 | 63 | Model evaluation · reasoning effort: max | Independent | Artificial Analysis |
| Artificial Analysis Coding Agent Index v1.3 | 67 | Claude Code · reasoning effort: xhigh | Independent | Artificial Analysis |
| DeepSWE 1.1 | 60 % | Claude Code · reasoning effort: xhigh | Independent | Artificial Analysis |
| Terminal-Bench 2.1 | 85 % | Claude Code · reasoning effort: xhigh | Independent | Artificial Analysis |
| SWE-Atlas-QnA 2026 | 55 % | Claude Code · reasoning effort: xhigh | Independent | Artificial Analysis |
| GDPval-AA v2 | 1861 Elo | Model evaluation · reasoning effort: max | Independent | Artificial Analysis |
| AA-Briefcase 2026-08 | 1720 Elo | Model evaluation · reasoning effort: max | Independent | Artificial Analysis |