GPT-4.1 is a closed-weights model from OpenAI, released April 14, 2025. It has an ECI score of 137 and ranks 125th of 249 tracked models. At the time of its release, it ranked 9th of 126 tracked models. At $2/Mtok for input tokens and $8/Mtok for output tokens, it is more expensive per token than similarly-capable models.
| Domain | Benchmark | ||||
|---|---|---|---|---|---|
ECI scoreBest score 166 | 137 | 137 | 140 | 138 | |
| Ranking | #125 | #126 | #107 | #119 | |
| Conceptual reasoning | ARC-AGI-2Best score 95.0% | ||||
ARC-AGI-1Best score 98.5% | |||||
| Games | Chess PuzzlesBest score 72.0% | ||||
| Mathematics | FrontierMath Tiers 1-3 (v2)Best score 93.7% | ||||
MATH Level 5Best score 98.1% | |||||
OTIS Mock AIME 2024-2025Best score 100.0% | |||||
| Science | Humanity's Last ExamBest score 46.5% | ||||
GPQA DiamondBest score 95.8% | |||||
| World knowledge | SimpleQA VerifiedBest score 75.6% |