Anthropic · stable
Anthropic's current Sonnet model, documented as a balance of speed and intelligence.
Specialist evidence
Source-native operational evidence for Claude Sonnet 5. Exact configurations remain separate. Costs are comparable only within the same selected benchmark/evaluation, and these rows never enter Overall Score.
| Evaluation | Exact configuration | Performance | Cost / task | Tokens / task | Execution |
|---|---|---|---|---|---|
| CursorBench · 3.2 · main CursorBench · checked 2026-08-29 | claude-sonnet-5-high Sonnet 5 High | 56.9% | $2.13 | 39483 | 57 steps |
| CursorBench · 3.2 · main CursorBench · checked 2026-08-29 | claude-sonnet-5-low Sonnet 5 Low | 47.7% | $0.870 | 16269 | 33 steps |
| CursorBench · 3.2 · main CursorBench · checked 2026-08-29 | claude-sonnet-5-max Sonnet 5 Max | 61.5% | $4.30 | 92882 | 86 steps |
| CursorBench · 3.2 · main CursorBench · checked 2026-08-29 | claude-sonnet-5-medium Sonnet 5 Medium | 52.4% | $1.44 | 26200 | 46 steps |
| CursorBench · 3.2 · main CursorBench · checked 2026-08-29 | claude-sonnet-5-xhigh Sonnet 5 Extra High | 58.7% | $2.77 | 52871 | 67 steps |
| Terminal-Bench 2.1 · 2.1 · verified Terminal-Bench 2.1 · checked 2026-08-29 | claude-sonnet-5-high Claude Code; reasoning high; Terminal-Bench 2.1 verified submission. | 74.6% | $3.24 | — | — |
