Anthropic · stable
Anthropic's most capable widely released model, documented for long-running agent workloads.
Specialist evidence
Source-native operational evidence for Claude Fable 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 |
|---|---|---|---|---|---|
| Artificial Analysis Coding Agents · 1.4 · composite Artificial Analysis Coding Agents · checked 2026-08-29 | claude-fable-5-max Claude Code - Fable 5 (max) (with fallback) | 67.2 | $11.69 | 13868660 | 137 steps |
| CursorBench · 3.2 · main CursorBench · checked 2026-08-29 | claude-fable-5-epoch-claude-fable-5-low Fable 5 Low | 62.1% | $4.46 | 18182 | 31 steps |
| CursorBench · 3.2 · main CursorBench · checked 2026-08-29 | claude-fable-5-high Fable 5 High | 66.5% | $8.77 | 43747 | 48 steps |
| CursorBench · 3.2 · main CursorBench · checked 2026-08-29 | claude-fable-5-max Fable 5 Max | 70.5% | $17.32 | 103525 | 72 steps |
| CursorBench · 3.2 · main CursorBench · checked 2026-08-29 | claude-fable-5-xhigh Fable 5 Extra High | 68.4% | $11.73 | 64971 | 56 steps |
| Terminal-Bench 2.1 · 2.1 · verified Terminal-Bench 2.1 · checked 2026-08-29 | claude-fable-5-high Terminus 2; reasoning high; Terminal-Bench 2.1 verified submission. | 80.4% | $4.93 | — | — |
| Terminal-Bench 2.1 · 2.1 · verified Terminal-Bench 2.1 · checked 2026-08-29 | claude-fable-5-xhigh Claude Code; reasoning xhigh; Terminal-Bench 2.1 verified submission. | 83.8% | $6.21 | — | — |
