Claude Opus 4.8
claude-opus
Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents
ReasoningTool CallingAttachments
Context Window
1M
Max Output
128K
Temperature
No
Open Weights
No
Knowledge Cutoff
N/A
Released
2026-05-28
Last Updated
2026-05-28
Modalities
Input:TextImagePDF
→Output:Text
Benchmarks
| Benchmark | Score | Metric | Harness |
|---|---|---|---|
| SWE-Bench Pro | 69.2 | resolve rate | - |
| Terminal-Bench | 74.6 | success rate | Terminus-2 |
Available Providers (2)
| Provider | Input /1M | Output /1M | Cache Read /1M | Cache Write /1M | Reasoning | Status |
|---|---|---|---|---|---|---|
| Rp 90.169 | Rp 450.843 | — | — | toggle, budget_tokens | — | |
| Rp 90.169 | Rp 450.843 | Rp 9.017 | Rp 112.711 | effort | — |