Model comparison
Claude Fable 5 vs GPT-5.4 mini
Claude Fable 5 is the stronger model overall, scoring 66.8 to 45.0 on the Noometry Index. GPT-5.4 mini costs 12× less per token, which makes it the better buy when Claude Fable 5's lead doesn't matter for your workload.
Last verified . 40 shared benchmarks.
Summary
- They share 40 benchmarks with published results for both. Claude Fable 5 scores higher in 10 categories and GPT-5.4 mini in 0 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Claude Fable 5 leads 76.8 to 30.4.
- The biggest single-benchmark swing is FrontierMath Tier 4: 90.2% for Claude Fable 5 and 9.8% for GPT-5.4 mini.
- GPT-5.4 mini is cheaper at $0.75 / $4.50 per million input/output tokens, against $10 / $50 for Claude Fable 5.
- Claude Fable 5 accepts more context: 1M tokens versus 400K.
Side by side
| Claude Fable 5 | GPT-5.4 mini | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 66.8 | 45.0 |
| Released | 2026-06-07 | 2026-03-17 |
| Weights | Proprietary | Proprietary |
| Context window | 1M | 400K |
| Max output | 128K | 128K |
| Input $ / M tokens | $10 | $0.75 |
| Output $ / M tokens | $50 | $4.50 |
| Results tracked | 62 | 46 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Fable 5 leads
Claude Fable 5: 70.6 (#4), GPT-5.4 mini: 45.2 (#72)
| Benchmark | Claude Fable 5 | GPT-5.4 mini |
|---|---|---|
| FrontierCode | 53.5% | 27% |
| LMArena WebDev | 1625 | 1397 |
| SciCode | 61% | 49.9% |
| WeirdML | 91.9% | 60.3% |
| LMArena Coding | 1519 | 1438 |
| ALE-Bench | 2,041 | 1,189 |
| DeepSWE | 69.9% | — |
| FrontierSWE | 47% | — |
| GSO | 78.4% | — |
| MirrorCode | 63.9% | — |
Agentic & Tool Use Claude Fable 5 leads
Claude Fable 5: 54.0 (#2), GPT-5.4 mini: 29.9 (#81)
| Benchmark | Claude Fable 5 | GPT-5.4 mini |
|---|---|---|
| APEX-Agents | 63.6% | — |
| Remote Labor Index | 16.1% | — |
| τ²-bench Banking | 39.7% | — |
| DeepResearch Bench | — | 36.3% |
| PostTrainBench | 41.8% | — |
| GBAEval | 74.5% | — |
| GDP.pdf | 30% | — |
| LMArena Search | 1230 | — |
| Vending-Bench 2 | 5,680 | — |
Reasoning Claude Fable 5 leads
Claude Fable 5: 76.8 (#6), GPT-5.4 mini: 30.4 (#85)
| Benchmark | Claude Fable 5 | GPT-5.4 mini |
|---|---|---|
| ARC-AGI-2 | 89.2% | 18.9% |
| Kagi LLM Benchmark | 91.4% | 37.9% |
| NYT Connections (extended) | 92.7% | 61.8% |
| ARC-AGI-1 | 98.5% | 63.7% |
| CritPt | 28.6% | 10% |
| Chess Puzzles | 41% | 24% |
| LMArena Hard Prompts | 1508 | 1424 |
| Mystery Game Puzzles | 52% | 11% |
| DTBench | 98.4% | 80% |
| LMCA | 61.1% | 40.8% |
| Epoch Capabilities Index | 162.06 | 148.84 |
| SimpleBench | 81.9% | — |
| EnigmaEval | 39.3% | — |
| Thematic Generalization | — | 61.7% |
| EBR-Bench | 39.5% | — |
| Surface Evolver Bench | 95% | — |
| Bench to the Future 3 | 0.13 | — |
| ForecastBench | — | 57 |
Math Claude Fable 5 leads
Claude Fable 5: 88.5 (#5), GPT-5.4 mini: 45.5 (#75)
| Benchmark | Claude Fable 5 | GPT-5.4 mini |
|---|---|---|
| FrontierMath (Tiers 1-3) | 87% | 51.2% |
| FrontierMath Tier 4 | 90.2% | 9.8% |
| OTIS Mock AIME 2024-2025 | 100% | 88.9% |
| ProofBench | 95% | 21% |
| LMArena Math | 1519 | 1419 |
| FrontierMath (Feb 2025 set) | — | 28.3% |
| FrontierMath Erdős | 0% | — |
| FrontierMath Tier 4 (v1) | — | 2.1% |
Knowledge Claude Fable 5 leads
Claude Fable 5: 62.2 (#25), GPT-5.4 mini: 51.5 (#67)
| Benchmark | Claude Fable 5 | GPT-5.4 mini |
|---|---|---|
| GPQA Diamond | 85.9% | 86.9% |
| SimpleQA Verified | 70.7% | 29.4% |
| LMArena Expert | 1534 | 1435 |
| Vectara Hallucination Rate | — | 5.5% |
Multimodal Claude Fable 5 leads
Claude Fable 5: 45.3 (#17), GPT-5.4 mini: 39.7 (#56)
| Benchmark | Claude Fable 5 | GPT-5.4 mini |
|---|---|---|
| LMArena Vision | 1324 | 1245 |
| Blueprint-Bench 2 | 38.6% | — |
| Furniture Assembly | 35.8% | — |
| LMArena Document | 1496 | — |
Multilingual Claude Fable 5 leads
Claude Fable 5: 57.3 (#9), GPT-5.4 mini: 51.9 (#96)
| Benchmark | Claude Fable 5 | GPT-5.4 mini |
|---|---|---|
| LMArena Non-English | 1481 | 1405 |
| LMArena Chinese | 1543 | 1446 |
| LMArena French | 1505 | 1440 |
| LMArena German | 1486 | 1409 |
| LMArena Japanese | 1506 | 1374 |
| LMArena Korean | 1488 | 1368 |
| LMArena Russian | 1504 | 1417 |
| LMArena Spanish | 1498 | 1405 |
Instruction Following Claude Fable 5 leads
Claude Fable 5: 78.6 (#8), GPT-5.4 mini: 74.1 (#102)
| Benchmark | Claude Fable 5 | GPT-5.4 mini |
|---|---|---|
| LMArena Instruction Following | 1502 | 1405 |
Long Context Claude Fable 5 leads
Claude Fable 5: 46.3 (#23), GPT-5.4 mini: 43.0 (#112)
| Benchmark | Claude Fable 5 | GPT-5.4 mini |
|---|---|---|
| LMArena Longer Query | 1509 | 1407 |
Writing & Preference Claude Fable 5 leads
Claude Fable 5: 75.9 (#5), GPT-5.4 mini: 64.0 (#58)
| Benchmark | Claude Fable 5 | GPT-5.4 mini |
|---|---|---|
| LMArena Text | 1491 | 1412 |
| LMArena Creative Writing | 1494 | 1370 |
| EQ-Bench Creative Writing | 1943 | 1665 |
| LMArena Multi-Turn | 1504 | 1429 |
| EQ-Bench 4 | 1340 | — |
Frequently asked questions
Is Claude Fable 5 better than GPT-5.4 mini?
Claude Fable 5 is the stronger model overall, scoring 66.8 to 45.0 on the Noometry Index. GPT-5.4 mini costs 12× less per token, which makes it the better buy when Claude Fable 5's lead doesn't matter for your workload.
Which is cheaper, Claude Fable 5 or GPT-5.4 mini?
GPT-5.4 mini is cheaper. It lists at $0.75 per million input tokens and $4.50 per million output tokens; Claude Fable 5 lists at $10 and $50.
Is Claude Fable 5 or GPT-5.4 mini better for coding?
Claude Fable 5 scores higher on coding benchmarks: 70.6 versus 45.2 in the Noometry coding category.
Which has the bigger context window?
Claude Fable 5 does, with 1M tokens against 400K.
How many benchmarks do Claude Fable 5 and GPT-5.4 mini share?
40 benchmarks have published results for both models. Claude Fable 5 has 62 scored results on Noometry and GPT-5.4 mini has 46.