Up to four models
Claude Fable 5 vs Nemotron 3 Ultra
On the index (GPT-4 = 100), Claude Fable 5 scores 183 and Nemotron 3 Ultra 141. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.
Claude Fable 5
183167–199
WeightsClosed
ReleasedJun 2026
Price in / out$10 / $50
Context1M
Core tests taken6 of 9
141124–157
WeightsOpen
ReleasedJun 2026
Price in / out$0.50 / $2.2
Context262K
Core tests taken3 of 9
Claude Fable 5 leads Nemotron 3 Ultra by 43 points, ahead on 3 of 3 shared core tests, at 22× the price.
Test by testthird-party results
Claude Fable 5
Nemotron 3 Ultra
Every result as a table
| Test | Claude Fable 5 | Nemotron 3 Ultra |
|---|---|---|
| Reasoning & knowledge | ||
| GPQA Diamond (Epoch AI) | 93.9% | 85.4% |
| GPQA Diamond (Vals AI) | 93.2% | 86.1% |
| MMLU Pro | 91.5% | 85.8% |
| SimpleBench | 81.9% | 41.7% |
| Maths | ||
| AIME-style maths (OTIS mock) | 100.0% | 86.7% |
| Coding | ||
| Code Migration | 55.1% | 4.9% |
| LiveCodeBench | 89.8% | 86.0% |
| ProgramBench | 2.0% | 0.0% |
| SWE-bench | 95.0% | 69.0% |
| Vibe Code Bench v1.1 | 90.3% | 7.6% |
| WeirdML | 91.9% | 43.5% |
| Agents | ||
| APEX-Agents | 63.6% | 22.7% |
| DTBench | 98.4% | 90.1% |
| LMCA | 61.2% | 36.9% |
| Mystery game puzzles | 52.0% | 20.0% |
| Professional work | ||
| Finance Agent (v2) | 56.3% | 37.7% |
| CorpFin v2 | 71.8% | 65.5% |
| EMB | 73.7% | 31.7% |
| LegalBench | 88.6% | 82.1% |
| Legal Research Bench | 49.5% | 15.4% |
| MedCode | 56.1% | 38.6% |
| TaxEval v2 | 76.9% | 73.1% |
Tests at least two of these models have taken. Bold rows are the core tests.