Model comparison
Claude Fable 5 vs o3
Claude Fable 5 is the stronger model overall, scoring 66.8 to 47.5 on the Noometry Index. o3 costs 5.7× less per token, which makes it the better buy when Claude Fable 5's lead doesn't matter for your workload.
Last verified . 38 shared benchmarks.
Summary
- They share 38 benchmarks with published results for both. Claude Fable 5 scores higher in 9 categories and o3 in 1 category; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Claude Fable 5 leads 76.8 to 32.0.
- The biggest single-benchmark swing is ARC-AGI-2: 89.2% for Claude Fable 5 and 6.5% for o3.
- o3 is cheaper at $2 / $8 per million input/output tokens, against $10 / $50 for Claude Fable 5.
- Claude Fable 5 accepts more context: 1M tokens versus 200K.
Side by side
| Claude Fable 5 | o3 | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 66.8 | 47.5 |
| Released | 2026-06-07 | 2025-04-16 |
| Weights | Proprietary | Proprietary |
| Context window | 1M | 200K |
| Max output | 128K | 100K |
| Input $ / M tokens | $10 | $2 |
| Output $ / M tokens | $50 | $8 |
| Results tracked | 62 | 63 |
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), o3: 46.8 (#64)
| Benchmark | Claude Fable 5 | o3 |
|---|---|---|
| GSO | 78.4% | 8.8% |
| WeirdML | 91.9% | 52.4% |
| LMArena Coding | 1519 | 1408 |
| ALE-Bench | 2,041 | 933.55 |
| SWE-bench Verified | — | 62.3% |
| DeepSWE | 69.9% | — |
| FrontierCode | 53.5% | — |
| SWE-bench Verified (bash only) | — | 58.4% |
| Aider Polyglot | — | 81.3% |
| LMArena WebDev | 1625 | — |
| FrontierSWE | 47% | — |
| SciCode | 61% | — |
| MirrorCode | 63.9% | — |
| CadEval | — | 74% |
Agentic & Tool Use Claude Fable 5 leads
Claude Fable 5: 54.0 (#2), o3: 34.5 (#44)
| Benchmark | Claude Fable 5 | o3 |
|---|---|---|
| LMArena Search | 1230 | 1144 |
| APEX-Agents | 63.6% | — |
| Berkeley Function Calling Leaderboard | — | 63% |
| GDPval | — | 30.8% |
| Remote Labor Index | 16.1% | — |
| τ²-bench Banking | 39.7% | — |
| DeepResearch Bench | — | 45.2% |
| OSWorld | — | 23% |
| PostTrainBench | 41.8% | — |
| GBAEval | 74.5% | — |
| GDP.pdf | 30% | — |
| METR Time Horizons | — | 65.4% |
| Vending-Bench 2 | 5,680 | — |
Reasoning Claude Fable 5 leads
Claude Fable 5: 76.8 (#6), o3: 32.0 (#78)
| Benchmark | Claude Fable 5 | o3 |
|---|---|---|
| ARC-AGI-2 | 89.2% | 6.5% |
| SimpleBench | 81.9% | 53.1% |
| Kagi LLM Benchmark | 91.4% | 67.6% |
| ARC-AGI-1 | 98.5% | 60.8% |
| CritPt | 28.6% | 1.4% |
| Chess Puzzles | 41% | 38% |
| EnigmaEval | 39.3% | 13.1% |
| LMArena Hard Prompts | 1508 | 1402 |
| Mystery Game Puzzles | 52% | 29% |
| DTBench | 98.4% | 84.8% |
| LMCA | 61.1% | 39.7% |
| Epoch Capabilities Index | 162.06 | 146.86 |
| NYT Connections (extended) | 92.7% | — |
| EBR-Bench | 39.5% | — |
| Surface Evolver Bench | 95% | — |
| Bench to the Future 3 | 0.13 | — |
| ForecastBench | — | 62.5 |
Math Claude Fable 5 leads
Claude Fable 5: 88.5 (#5), o3: 50.2 (#58)
| Benchmark | Claude Fable 5 | o3 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 87% | 33.3% |
| OTIS Mock AIME 2024-2025 | 100% | 84.4% |
| LMArena Math | 1519 | 1426 |
| FrontierMath Tier 4 | 90.2% | — |
| ProofBench | 95% | — |
| Omni-MATH | — | 71.4% |
| MATH Level 5 | — | 97.8% |
| FrontierMath (Feb 2025 set) | — | 18.7% |
| FrontierMath Erdős | 0% | — |
| FrontierMath Tier 4 (v1) | — | 2.1% |
Knowledge Claude Fable 5 leads
Claude Fable 5: 62.2 (#25), o3: 54.6 (#52)
| Benchmark | Claude Fable 5 | o3 |
|---|---|---|
| GPQA Diamond | 85.9% | 81.8% |
| SimpleQA Verified | 70.7% | 49.4% |
| LMArena Expert | 1534 | 1402 |
| Humanity's Last Exam | — | 20.3% |
| MMLU-Pro | — | 85.9% |
| Confabulations | — | 14.4% |
| GPQA (HELM) | — | 75.3% |
Multimodal Claude Fable 5 leads
Claude Fable 5: 45.3 (#17), o3: 41.4 (#36)
| Benchmark | Claude Fable 5 | o3 |
|---|---|---|
| LMArena Vision | 1324 | 1214 |
| GeoBench | — | 74% |
| VPCT | — | 52% |
| Blueprint-Bench 2 | 38.6% | — |
| Furniture Assembly | 35.8% | — |
| LMArena Document | 1496 | — |
Multilingual Claude Fable 5 leads
Claude Fable 5: 57.3 (#9), o3: 51.7 (#105)
| Benchmark | Claude Fable 5 | o3 |
|---|---|---|
| LMArena Non-English | 1481 | 1401 |
| LMArena Chinese | 1543 | 1437 |
| LMArena French | 1505 | 1430 |
| LMArena German | 1486 | 1420 |
| LMArena Japanese | 1506 | 1403 |
| LMArena Korean | 1488 | 1370 |
| LMArena Russian | 1504 | 1406 |
| LMArena Spanish | 1498 | 1395 |
Instruction Following Claude Fable 5 leads
Claude Fable 5: 78.6 (#8), o3: 72.8 (#127)
| Benchmark | Claude Fable 5 | o3 |
|---|---|---|
| LMArena Instruction Following | 1502 | 1368 |
| IFEval | — | 86.9% |
Long Context o3 leads
Claude Fable 5: 46.3 (#23), o3: 53.3 (#6)
| Benchmark | Claude Fable 5 | o3 |
|---|---|---|
| LMArena Longer Query | 1509 | 1372 |
| Fiction.LiveBench | — | 88.9% |
| CL-bench | — | 17.8% |
Writing & Preference Claude Fable 5 leads
Claude Fable 5: 75.9 (#5), o3: 63.5 (#64)
| Benchmark | Claude Fable 5 | o3 |
|---|---|---|
| LMArena Text | 1491 | 1410 |
| LMArena Creative Writing | 1494 | 1359 |
| EQ-Bench Creative Writing | 1943 | 1676 |
| LMArena Multi-Turn | 1504 | 1405 |
| Short-Story Creative Writing | — | 83.9% |
| WildBench | — | 86.1% |
| EQ-Bench 4 | 1340 | — |
Frequently asked questions
Is Claude Fable 5 better than o3?
Claude Fable 5 is the stronger model overall, scoring 66.8 to 47.5 on the Noometry Index. o3 costs 5.7× 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 o3?
o3 is cheaper. It lists at $2 per million input tokens and $8 per million output tokens; Claude Fable 5 lists at $10 and $50.
Is Claude Fable 5 or o3 better for coding?
Claude Fable 5 scores higher on coding benchmarks: 70.6 versus 46.8 in the Noometry coding category.
Which has the bigger context window?
Claude Fable 5 does, with 1M tokens against 200K.
How many benchmarks do Claude Fable 5 and o3 share?
38 benchmarks have published results for both models. Claude Fable 5 has 62 scored results on Noometry and o3 has 63.