Claude Fable 5 vs Claude Fable 5.2
Claude Fable 5.2 leads Claude Fable 5 by 12.3 points on the Shogo Agent Index. Claude Fable 5 wins 1 and Claude Fable 5.2 wins 18 of 20 shared benchmarks. Claude Fable 5.2 is cheaper for agent workloads.
Add more models to this comparison →Pillars
| Pillar | Claude Fable 5 | Claude Fable 5.2 |
|---|---|---|
| Agentic coding | 74 | 86 |
| Long-horizon autonomy | 69 | 78 |
| Knowledge-work agents | 52 | 67 |
| Computer use | 55 | 77 |
| Security & context learning | 59 | 65 |
Shared benchmarks
| Benchmark | Claude Fable 5 | Claude Fable 5.2 |
|---|---|---|
| APEX-Agentsindex | 63.6% | 80.1% |
| Vending-Bench 2index | $5,680 | $9,628 |
| FrontierSWEindex | 47.0% | 69.9% |
| Remote Labor Indexindex | 16.1% | 41.3% |
| PostTrainBenchindex | 37.3% | 44.6% |
| ARC-AGI | 98.5% | 98.5% |
| ARC-AGI-2 | 89.2% | 93.8% |
| Chess Puzzles | 47.0% | 79.0% |
| CritPt | 28.6% | 35.7% |
| DTBench | 98.4% | 99.2% |
| FrontierMath Tier 4 | 90.2% | 100.0% |
| FrontierMath Tiers 1-3 | 87.0% | 95.1% |
| GPQA Diamond | 93.9% | 96.0% |
| LMCA | 61.2% | 70.6% |
| OTIS Mock AIME | 100.0% | 99.7% |
| ProofBench | 95.0% | 100.0% |
| SciCode | 61.0% | 67.7% |
| SimpleBench | 81.9% | 94.9% |
| SimpleQA Verified | 70.7% | 94.0% |
| WeirdML v3 | 15.6% | 43.1% |
Specs and price
| Claude Fable 5 | Claude Fable 5.2 | |
|---|---|---|
| Agent Index | 64.1 (#12) | 76.4 (#1) |
| Agent $/M | $4.62 | $3.21 |
| Input / output $/M | $10.00 / $50.00 | $8.00 / $40.00 |
| Context | 1M | 1M |
| Released | Jun 9, 2026 | Nov 9, 2026 |
| Weights | Closed | Closed |
| API hosts | 4 | 4 |
Don’t pick one. Route.
Shogo can run Claude Fable 5 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