Claude Fable 5.2 vs Claude Sonnet 5.5
Claude Fable 5.2 leads Claude Sonnet 5.5 by 7.4 points on the Shogo Agent Index, but their ranges overlap, so treat it as close. Claude Fable 5.2 wins 16 and Claude Sonnet 5.5 wins 2 of 19 shared benchmarks. Claude Sonnet 5.5 is cheaper for agent workloads.
Add more models to this comparison →Pillars
| Pillar | Claude Fable 5.2 | Claude Sonnet 5.5 |
|---|---|---|
| Agentic coding | 86 | 79 |
| Long-horizon autonomy | 78 | 75 |
| Knowledge-work agents | 67 | 57 |
| Computer use | 77 | 61 |
| Security & context learning | 65 | 61 |
Shared benchmarks
| Benchmark | Claude Fable 5.2 | Claude Sonnet 5.5 |
|---|---|---|
| APEX-Agentsindex | 80.1% | 75.5% |
| Vending-Bench 2index | $9,628 | $11,165 |
| FrontierSWEindex | 69.9% | 61.9% |
| CursorBenchindex | 60.0% | 55.5% |
| ARC-AGI | 98.5% | 96.5% |
| ARC-AGI-2 | 93.8% | 83.9% |
| Chess Puzzles | 79.0% | 59.0% |
| CritPt | 35.7% | 31.4% |
| DTBench | 99.2% | 97.3% |
| FrontierMath Tier 4 | 100.0% | 80.5% |
| FrontierMath Tiers 1-3 | 95.1% | 88.8% |
| GPQA Diamond | 96.0% | 95.6% |
| LMCA | 70.6% | 68.3% |
| OTIS Mock AIME | 99.7% | 100.0% |
| ProofBench | 100.0% | 100.0% |
| SciCode | 67.7% | 61.0% |
| SimpleBench | 94.9% | 75.9% |
| SimpleQA Verified | 94.0% | 46.5% |
| WeirdML v3 | 43.1% | 19.9% |
Specs and price
| Claude Fable 5.2 | Claude Sonnet 5.5 | |
|---|---|---|
| Agent Index | 76.4 (#1) | 69 (#9) |
| Agent $/M | $3.21 | $0.84 |
| Input / output $/M | $8.00 / $40.00 | $2.00 / $10.00 |
| Context | 1M | 1M |
| Released | Nov 9, 2026 | Sep 28, 2026 |
| Weights | Closed | Closed |
| API hosts | 4 | 5 |
Don’t pick one. Route.
Shogo can run Claude Fable 5.2 and Claude Sonnet 5.5 side by side and send each step of an agent to the one that’s cheaper for it.
Try both in Shogo