Claude Fable 5 vs GPT-5.4
Claude Fable 5 leads GPT-5.4 by 12.5 points on the Shogo Agent Index. Claude Fable 5 wins 18 and GPT-5.4 wins 2 of 20 shared benchmarks. GPT-5.4 is cheaper for agent workloads.
Add more models to this comparison →Pillars
| Pillar | Claude Fable 5 | GPT-5.4 |
|---|---|---|
| Agentic coding | 74 | 57 |
| Long-horizon autonomy | 69 | 61 |
| Knowledge-work agents | 52 | 40 |
| Computer use | 55 | 42 |
| Security & context learning | 59 | 54 |
Shared benchmarks
| Benchmark | Claude Fable 5 | GPT-5.4 |
|---|---|---|
| APEX-Agentsindex | 63.6% | 52.4% |
| DeepSWEindex | 69.9% | 51.8% |
| Vending-Bench 2index | $5,680 | $6,144 |
| GSO-Benchindex | 78.4% | 31.4% |
| PostTrainBenchindex | 37.3% | 17.8% |
| ARC-AGI | 98.5% | 93.7% |
| ARC-AGI-2 | 89.2% | 74.0% |
| Chess Puzzles | 47.0% | 48.0% |
| CritPt | 28.6% | 23.4% |
| DTBench | 98.4% | 94.4% |
| FrontierMath Tier 4 | 90.2% | 49.0% |
| FrontierMath Tiers 1-3 | 87.0% | 78.6% |
| GPQA Diamond | 93.9% | 93.3% |
| LMCA | 61.2% | 52.0% |
| MirrorCode | 70.0% | 23.3% |
| OTIS Mock AIME | 100.0% | 97.8% |
| ProofBench | 95.0% | 56.0% |
| SciCode | 61.0% | 56.6% |
| SimpleQA Verified | 70.7% | 45.1% |
| WebDev Arena | 1628 | 1463 |
Specs and price
| Claude Fable 5 | GPT-5.4 | |
|---|---|---|
| Agent Index | 64.1 (#12) | 51.6 (#33) |
| Agent $/M | $4.62 | $1.27 |
| Input / output $/M | $10.00 / $50.00 | $2.50 / $15.00 |
| Context | 1M | 1.1M |
| Released | Jun 9, 2026 | Mar 5, 2026 |
| Weights | Closed | Closed |
| API hosts | 4 | 4 |
Don’t pick one. Route.
Shogo can run Claude Fable 5 and GPT-5.4 side by side and send each step of an agent to the one that’s cheaper for it.
Try both in Shogo