Up to four models
o4-mini vs GPT-5.4 Nano
On the index (GPT-4 = 100), o4-mini scores 141 and GPT-5.4 Nano 141. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.
GPT-5.4 Nano
141125–157
WeightsClosed
ReleasedMar 2026
Price in / out$0.20 / $1.25
Context400K
Core tests taken4 of 9
o4-mini and GPT-5.4 Nano are too close to call (0 points apart, within the uncertainty), ahead on 2 of 3 shared core tests.
Test by testthird-party results
o4-miniGPT-5.4 Nano
Every result as a table
| Test | o4-mini | GPT-5.4 Nano |
|---|---|---|
| Reasoning & knowledge | ||
| GPQA Diamond (Epoch AI) | 79.6% | 78.5% |
| GPQA Diamond (Vals AI) | 74.5% | 77.5% |
| MMLU Pro | 80.6% | 77.2% |
| MMMU Pro | 79.7% | 73.6% |
| SimpleQA Verified | 19.6% | 11.7% |
| Maths | ||
| FrontierMath (tiers 1-3) | 36.1% | 44.9% |
| AIME-style maths (OTIS mock) | 81.7% | 87.8% |
| Chess puzzles | 22.0% | 45.0% |
| FrontierMath (2025 set) | 24.8% | 25.9% |
| FrontierMath tier 4 (2025 set) | 6.3% | 6.3% |
| FrontierMath (tier 4) | 4.9% | 12.2% |
| Coding | ||
| LiveCodeBench | 82.2% | 84.0% |
| WeirdML | 52.6% | 49.2% |
| Agents | ||
| DTBench | 77.6% | 80.3% |
| LMCA | 26.5% | 36.9% |
| Mystery game puzzles | 5.0% | 9.0% |
| Professional work | ||
| CorpFin v2 | 59.0% | 61.2% |
| LegalBench | 79.2% | 77.9% |
| MedCode | 33.8% | 41.0% |
| MedScribe | 69.1% | 77.1% |
| MortgageTax | 64.8% | 59.1% |
| TaxEval v2 | 74.8% | 67.4% |
| Novel problems | ||
| ARC-AGI-2 | 6.1% | 5.7% |
| ARC-AGI-1 | 58.7% | 51.5% |
Task length · METR38 sec
Tests at least two of these models have taken. Bold rows are the core tests.