Model comparison
Claude Fable 5 vs GPT-4
Claude Fable 5 is the stronger model overall, scoring 66.8 to 29.1 on the Noometry Index.
Last verified . 26 shared benchmarks.
Summary
- They share 26 benchmarks with published results for both. Claude Fable 5 scores higher in 8 categories and GPT-4 in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Claude Fable 5 leads 88.5 to 10.8.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 100% for Claude Fable 5 and 1.1% for GPT-4.
- Claude Fable 5 is cheaper at $10 / $50 per million input/output tokens, against $30 / $60 for GPT-4.
- Claude Fable 5 accepts more context: 1M tokens versus 8K.
Side by side
| Claude Fable 5 | GPT-4 | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 66.8 | 29.1 |
| Released | 2026-06-07 | 2023-03-14 |
| Weights | Proprietary | Proprietary |
| Context window | 1M | 8K |
| Max output | 128K | 8K |
| Input $ / M tokens | $10 | $30 |
| Output $ / M tokens | $50 | $60 |
| Results tracked | 62 | 38 |
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-4: 31.6 (#283)
| Benchmark | Claude Fable 5 | GPT-4 |
|---|---|---|
| WeirdML | 91.9% | 12.4% |
| LMArena Coding | 1519 | 1254 |
| DeepSWE | 69.9% | — |
| FrontierCode | 53.5% | — |
| LMArena WebDev | 1625 | — |
| FrontierSWE | 47% | — |
| SciCode | 61% | — |
| GSO | 78.4% | — |
| BigCodeBench Instruct | — | 46% |
| MirrorCode | 63.9% | — |
| BigCodeBench Complete | — | 57.2% |
| ALE-Bench | 2,041 | — |
| HumanEval+ | — | 79.3% |
Agentic & Tool Use Not comparable
Claude Fable 5: 54.0 (#2), GPT-4: —
| Benchmark | Claude Fable 5 | GPT-4 |
|---|---|---|
| APEX-Agents | 63.6% | — |
| Remote Labor Index | 16.1% | — |
| τ²-bench Banking | 39.7% | — |
| PostTrainBench | 41.8% | — |
| GBAEval | 74.5% | — |
| GDP.pdf | 30% | — |
| LMArena Search | 1230 | — |
| METR Time Horizons | — | 36.1% |
| Vending-Bench 2 | 5,680 | — |
Reasoning Claude Fable 5 leads
Claude Fable 5: 76.8 (#6), GPT-4: 17.8 (#289)
| Benchmark | Claude Fable 5 | GPT-4 |
|---|---|---|
| Chess Puzzles | 41% | 4% |
| LMArena Hard Prompts | 1508 | 1241 |
| Mystery Game Puzzles | 52% | 12% |
| DTBench | 98.4% | 62.7% |
| LMCA | 61.1% | 17.1% |
| Epoch Capabilities Index | 162.06 | 125.89 |
| ARC-AGI-2 | 89.2% | — |
| SimpleBench | 81.9% | — |
| Kagi LLM Benchmark | 91.4% | — |
| NYT Connections (extended) | 92.7% | — |
| ARC-AGI-1 | 98.5% | — |
| CritPt | 28.6% | — |
| EnigmaEval | 39.3% | — |
| EBR-Bench | 39.5% | — |
| Surface Evolver Bench | 95% | — |
| Bench to the Future 3 | 0.13 | — |
| BIG-Bench Hard | — | 75.1% |
| ForecastBench | — | 57.8 |
| HellaSwag | — | 95.3% |
| WinoGrande | — | 87.5% |
Math Claude Fable 5 leads
Claude Fable 5: 88.5 (#5), GPT-4: 10.8 (#309)
| Benchmark | Claude Fable 5 | GPT-4 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 100% | 1.1% |
| LMArena Math | 1519 | 1269 |
| FrontierMath (Tiers 1-3) | 87% | — |
| FrontierMath Tier 4 | 90.2% | — |
| ProofBench | 95% | — |
| MATH Level 5 | — | 23% |
| FrontierMath Erdős | 0% | — |
| GSM8K | — | 92% |
Knowledge Claude Fable 5 leads
Claude Fable 5: 62.2 (#25), GPT-4: 18.4 (#282)
| Benchmark | Claude Fable 5 | GPT-4 |
|---|---|---|
| GPQA Diamond | 85.9% | 35.7% |
| LMArena Expert | 1534 | 1211 |
| SimpleQA Verified | 70.7% | — |
| MMLU | — | 86.4% |
| TriviaQA | — | 84.8% |
Multimodal Not comparable
Claude Fable 5: 45.3 (#17), GPT-4: —
| Benchmark | Claude Fable 5 | GPT-4 |
|---|---|---|
| LMArena Vision | 1324 | — |
| Blueprint-Bench 2 | 38.6% | — |
| Furniture Assembly | 35.8% | — |
| LMArena Document | 1496 | — |
Multilingual Claude Fable 5 leads
Claude Fable 5: 57.3 (#9), GPT-4: 40.6 (#215)
| Benchmark | Claude Fable 5 | GPT-4 |
|---|---|---|
| LMArena Non-English | 1481 | 1246 |
| LMArena Chinese | 1543 | 1242 |
| LMArena French | 1505 | 1283 |
| LMArena German | 1486 | 1251 |
| LMArena Japanese | 1506 | 1209 |
| LMArena Korean | 1488 | 1184 |
| LMArena Russian | 1504 | 1251 |
| LMArena Spanish | 1498 | 1261 |
Instruction Following Claude Fable 5 leads
Claude Fable 5: 78.6 (#8), GPT-4: 65.3 (#222)
| Benchmark | Claude Fable 5 | GPT-4 |
|---|---|---|
| LMArena Instruction Following | 1502 | 1241 |
Long Context Claude Fable 5 leads
Claude Fable 5: 46.3 (#23), GPT-4: 37.7 (#212)
| Benchmark | Claude Fable 5 | GPT-4 |
|---|---|---|
| LMArena Longer Query | 1509 | 1244 |
Writing & Preference Claude Fable 5 leads
Claude Fable 5: 75.9 (#5), GPT-4: 34.9 (#268)
| Benchmark | Claude Fable 5 | GPT-4 |
|---|---|---|
| LMArena Text | 1491 | 1263 |
| LMArena Creative Writing | 1494 | 1244 |
| EQ-Bench Creative Writing | 1943 | 752 |
| LMArena Multi-Turn | 1504 | 1257 |
| EQ-Bench 4 | 1340 | — |
Frequently asked questions
Is Claude Fable 5 better than GPT-4?
Claude Fable 5 is the stronger model overall, scoring 66.8 to 29.1 on the Noometry Index.
Which is cheaper, Claude Fable 5 or GPT-4?
Claude Fable 5 is cheaper. It lists at $10 per million input tokens and $50 per million output tokens; GPT-4 lists at $30 and $60.
Is Claude Fable 5 or GPT-4 better for coding?
Claude Fable 5 scores higher on coding benchmarks: 70.6 versus 31.6 in the Noometry coding category.
Which has the bigger context window?
Claude Fable 5 does, with 1M tokens against 8K.
How many benchmarks do Claude Fable 5 and GPT-4 share?
26 benchmarks have published results for both models. Claude Fable 5 has 62 scored results on Noometry and GPT-4 has 38.