Claude Fable 5.1 vs Claude Fable 5.2
Claude Fable 5.2 leads Claude Fable 5.1 by 8.2 points on the Shogo Agent Index, but their ranges overlap, so treat it as close. Claude Fable 5.1 wins 2 and Claude Fable 5.2 wins 17 of 20 shared benchmarks. Claude Fable 5.2 is cheaper for agent workloads.
Add more models to this comparison →Pillars
| Pillar | Claude Fable 5.1 | Claude Fable 5.2 |
|---|---|---|
| Agentic coding | 80 | 86 |
| Long-horizon autonomy | 71 | 78 |
| Knowledge-work agents | 55 | 67 |
| Computer use | 62 | 77 |
| Security & context learning | 61 | 65 |
Shared benchmarks
| Benchmark | Claude Fable 5.1 | Claude Fable 5.2 |
|---|---|---|
| APEX-Agentsindex | 68.6% | 80.1% |
| Vending-Bench 2index | $5,422 | $9,628 |
| FrontierSWEindex | 56.3% | 69.9% |
| Remote Labor Indexindex | 17.9% | 41.3% |
| CursorBenchindex | 51.8% | 60.0% |
| PostTrainBenchindex | 44.6% | 44.6% |
| ARC-AGI | 97.5% | 98.5% |
| ARC-AGI-2 | 90.0% | 93.8% |
| Chess Puzzles | 48.0% | 79.0% |
| CritPt | 31.1% | 35.7% |
| DTBench | 97.6% | 99.2% |
| FrontierMath Tier 4 | 87.8% | 100.0% |
| FrontierMath Tiers 1-3 | 90.2% | 95.1% |
| LMCA | 65.5% | 70.6% |
| OTIS Mock AIME | 100.0% | 99.7% |
| ProofBench | 100.0% | 100.0% |
| SciCode | 63.1% | 67.7% |
| SimpleBench | 86.6% | 94.9% |
| SimpleQA Verified | 70.8% | 94.0% |
| WeirdML v3 | 26.0% | 43.1% |
Specs and price
| Claude Fable 5.1 | Claude Fable 5.2 | |
|---|---|---|
| Agent Index | 68.2 (#10) | 76.4 (#1) |
| Agent $/M | $4.01 | $3.21 |
| Input / output $/M | $10.00 / $50.00 | $8.00 / $40.00 |
| Context | 1M | 1M |
| Released | Sep 1, 2026 | Nov 9, 2026 |
| Weights | Closed | Closed |
| API hosts | 3 | 4 |
Don’t pick one. Route.
Shogo can run Claude Fable 5.1 and Claude Fable 5.2 side by side and send each step of an agent to the one that’s cheaper for it.
Try both in Shogo