Model comparison
Claude Fable 5.1 vs GPT-4
Claude Fable 5.1 is the stronger model overall, scoring 69.0 to 29.1 on the Noometry Index.
Last verified . 25 shared benchmarks.
Summary
- They share 25 benchmarks with published results for both. Claude Fable 5.1 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 5.1 leads 89.6 to 10.8.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 100% for Claude Fable 5.1 and 1.1% for GPT-4.
- Claude Fable 5.1 is cheaper at $10 / $50 per million input/output tokens, against $30 / $60 for GPT-4.
- Claude Fable 5.1 accepts more context: 1M tokens versus 8K.
Side by side
| Claude Fable 5.1 | GPT-4 | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 69.0 | 29.1 |
| Released | 2026-09-01 | 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 | 52 | 38 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Fable 5.1 leads
Claude Fable 5.1: 74.7 (#1), GPT-4: 31.6 (#283)
| Benchmark | Claude Fable 5.1 | GPT-4 |
|---|---|---|
| WeirdML | 92.9% | 12.4% |
| LMArena Coding | 1528 | 1254 |
| FrontierCode | 50.9% | — |
| CursorBench | 51.8% | — |
| LMArena WebDev | 1744 | — |
| FrontierSWE | 56.3% | — |
| SciCode | 63.1% | — |
| GSO | 88.2% | — |
| BigCodeBench Instruct | — | 46% |
| MirrorCode | 73.3% | — |
| BigCodeBench Complete | — | 57.2% |
| ALE-Bench | 2,143 | — |
| HumanEval+ | — | 79.3% |
Agentic & Tool Use Not comparable
Claude Fable 5.1: 50.7 (#5), GPT-4: —
| Benchmark | Claude Fable 5.1 | GPT-4 |
|---|---|---|
| APEX-Agents | 68.6% | — |
| Remote Labor Index | 17.9% | — |
| GDP.pdf | 29.6% | — |
| METR Time Horizons | — | 36.1% |
| Vending-Bench 2 | 5,422 | — |
Reasoning Claude Fable 5.1 leads
Claude Fable 5.1: 76.7 (#7), GPT-4: 17.8 (#289)
| Benchmark | Claude Fable 5.1 | GPT-4 |
|---|---|---|
| Chess Puzzles | 47% | 4% |
| LMArena Hard Prompts | 1526 | 1241 |
| Mystery Game Puzzles | 58% | 12% |
| DTBench | 97.6% | 62.7% |
| LMCA | 65.5% | 17.1% |
| Epoch Capabilities Index | 164.7 | 125.89 |
| ARC-AGI-2 | 90% | — |
| NYT Connections (extended) | 90% | — |
| ARC-AGI-1 | 97.5% | — |
| CritPt | 31.1% | — |
| EBR-Bench | 57.1% | — |
| BIG-Bench Hard | — | 75.1% |
| ForecastBench | — | 57.8 |
| HellaSwag | — | 95.3% |
| WinoGrande | — | 87.5% |
Math Claude Fable 5.1 leads
Claude Fable 5.1: 89.6 (#4), GPT-4: 10.8 (#309)
| Benchmark | Claude Fable 5.1 | GPT-4 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 100% | 1.1% |
| LMArena Math | 1525 | 1269 |
| FrontierMath (Tiers 1-3) | 90.2% | — |
| FrontierMath Tier 4 | 87.8% | — |
| ProofBench | 100% | — |
| MATH Level 5 | — | 23% |
| FrontierMath Erdős | 0% | — |
| GSM8K | — | 92% |
Knowledge Claude Fable 5.1 leads
Claude Fable 5.1: 69.6 (#6), GPT-4: 18.4 (#282)
| Benchmark | Claude Fable 5.1 | GPT-4 |
|---|---|---|
| LMArena Expert | 1535 | 1211 |
| GPQA Diamond | — | 35.7% |
| Humanity's Last Exam | 46.5% | — |
| SimpleQA Verified | 70.8% | — |
| MMLU | — | 86.4% |
| TriviaQA | — | 84.8% |
Multimodal Not comparable
Claude Fable 5.1: 53.9 (#4), GPT-4: —
| Benchmark | Claude Fable 5.1 | GPT-4 |
|---|---|---|
| LMArena Vision | 1318 | — |
| Blueprint-Bench 2 | 41.9% | — |
| Furniture Assembly | 70% | — |
| LMArena Document | 1513 | — |
Multilingual Claude Fable 5.1 leads
Claude Fable 5.1: 59.1 (#3), GPT-4: 40.6 (#215)
| Benchmark | Claude Fable 5.1 | GPT-4 |
|---|---|---|
| LMArena Non-English | 1507 | 1246 |
| LMArena Chinese | 1586 | 1242 |
| LMArena French | 1525 | 1283 |
| LMArena German | 1500 | 1251 |
| LMArena Japanese | 1543 | 1209 |
| LMArena Korean | 1534 | 1184 |
| LMArena Russian | 1521 | 1251 |
| LMArena Spanish | 1516 | 1261 |
Instruction Following Claude Fable 5.1 leads
Claude Fable 5.1: 79.2 (#6), GPT-4: 65.3 (#222)
| Benchmark | Claude Fable 5.1 | GPT-4 |
|---|---|---|
| LMArena Instruction Following | 1517 | 1241 |
Long Context Claude Fable 5.1 leads
Claude Fable 5.1: 46.7 (#20), GPT-4: 37.7 (#212)
| Benchmark | Claude Fable 5.1 | GPT-4 |
|---|---|---|
| LMArena Longer Query | 1522 | 1244 |
Writing & Preference Claude Fable 5.1 leads
Claude Fable 5.1: 79.2 (#2), GPT-4: 34.9 (#268)
| Benchmark | Claude Fable 5.1 | GPT-4 |
|---|---|---|
| LMArena Text | 1510 | 1263 |
| LMArena Creative Writing | 1507 | 1244 |
| EQ-Bench Creative Writing | 2162 | 752 |
| LMArena Multi-Turn | 1492 | 1257 |
Frequently asked questions
Is Claude Fable 5.1 better than GPT-4?
Claude Fable 5.1 is the stronger model overall, scoring 69.0 to 29.1 on the Noometry Index.
Which is cheaper, Claude Fable 5.1 or GPT-4?
Claude Fable 5.1 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 5.1 or GPT-4 better for coding?
Claude Fable 5.1 scores higher on coding benchmarks: 74.7 versus 31.6 in the Noometry coding category.
Which has the bigger context window?
Claude Fable 5.1 does, with 1M tokens against 8K.
How many benchmarks do Claude Fable 5.1 and GPT-4 share?
25 benchmarks have published results for both models. Claude Fable 5.1 has 52 scored results on Noometry and GPT-4 has 38.