Claude Fable 5 vs Claude Sonnet 5
Claude Fable 5 leads Claude Sonnet 5 by 11.7 points on the Shogo Agent Index, but their ranges overlap, so treat it as close. Claude Fable 5 wins 19 and Claude Sonnet 5 wins 1 of 20 shared benchmarks. Claude Sonnet 5 is cheaper for agent workloads.
Add more models to this comparison →Pillars
| Pillar | Claude Fable 5 | Claude Sonnet 5 |
|---|---|---|
| Agentic coding | 74 | 57 |
| Long-horizon autonomy | 69 | 65 |
| Knowledge-work agents | 52 | 43 |
| Computer use | 55 | 40 |
| Security & context learning | 59 | 51 |
Shared benchmarks
| Benchmark | Claude Fable 5 | Claude Sonnet 5 |
|---|---|---|
| APEX-Agentsindex | 63.6% | 54.5% |
| DeepSWEindex | 69.9% | 53.8% |
| Vending-Bench 2index | $5,680 | $6,378 |
| GSO-Benchindex | 78.4% | 37.3% |
| Chess Puzzles | 47.0% | 45.0% |
| CritPt | 28.6% | 16.9% |
| DTBench | 98.4% | 92.5% |
| FrontierCode | 53.5% | 42.7% |
| FrontierMath Tier 4 | 90.2% | 29.3% |
| FrontierMath Tiers 1-3 | 87.0% | 65.6% |
| GeoBench | 89.0% | 74.0% |
| GPQA Diamond | 93.9% | 90.5% |
| LMCA | 61.2% | 50.0% |
| MirrorCode | 70.0% | 63.3% |
| OTIS Mock AIME | 100.0% | 94.7% |
| ProofBench | 95.0% | 77.0% |
| SciCode | 61.0% | 54.3% |
| SimpleBench | 81.9% | 60.6% |
| SimpleQA Verified | 70.7% | 33.7% |
| WebDev Arena | 1628 | 1537 |
Specs and price
| Claude Fable 5 | Claude Sonnet 5 | |
|---|---|---|
| Agent Index | 64.1 (#12) | 52.4 (#32) |
| Agent $/M | $4.62 | $0.92 |
| Input / output $/M | $10.00 / $50.00 | $2.00 / $10.00 |
| Context | 1M | 1M |
| Released | Jun 9, 2026 | Jun 30, 2026 |
| Weights | Closed | Closed |
| API hosts | 4 | 5 |
Don’t pick one. Route.
Shogo can run Claude Fable 5 and Claude Sonnet 5 side by side and send each step of an agent to the one that’s cheaper for it.
Try both in Shogo