OpenAI
availableShows if the model has enough results for an index.GPT-4.1 mini
GPT-4.1 mini is a non-reasoning model from OpenAI in the GPT-4.1 family. 28 benchmarks count toward its score, in 7 categories.
IndexOverall score out of 100.34.8 ±3.6
CoverageShare of the index weight with results.95%
SpeedOutput tokens per second.38/s
Input / 1MUS dollars per 1M input tokens.$0.4 batch $0.2
Output / 1MUS dollars per 1M output tokens.$1.6 batch $0.8 US dollars per 1M output tokens in a batch.
ContextMaximum tokens in one request.1.05M
EloLMArena rating and rank.1340 (#199)
The index is a score out of 100. The ± range shows how much it can change. Batch work costs less.
38,631 votes. Elo shows what people prefer. It does not change the score.
CapabilitiesScore per category, out of 100.
Out of 100Results
28 counted| BenchmarkThe test name. | CategoryThe capability that the test measures. | ResultThe score from the publisher. | IndexThis result as a score out of 100. | RunThe settings of the run. | DateDate of the result. | Published byThe source of the result. |
|---|---|---|---|---|---|---|
| Instruction-Following Eval | Instruction | 88.5% | 42.0 | — | — | Jeffrey Zhou et al. |
| MATH 500 | Math | 88.0% | 40.2 | — | 9 Jan 2026 | Vals AI |
| MGSM | Multilingual | 87.8% | — | — | 9 Jan 2026 | Vals AI |
| Massive Multitask Language Understanding | Knowledge | 87.5% | — | — | — | Dan Hendrycks et al. |
| MATH level 5 | Math | 87.3% | 41.6 | — | — | Epoch AI |
| MMLU Pro | Knowledge | 77.2% | 42.2 | — | 1 Sept 2026 | Vals AI |
| MMMU Pro | Multimodal | 70.5% | 42.1 | — | 1 Sept 2026 | Vals AI |
| GPQA Diamond | Knowledge | 67.9% | 41.0 | — | 1 Sept 2026 | Vals AI |
| Artificial Analysis GPQA Diamond | Knowledge | 66.4% | 36.6 | — | — | Artificial Analysis |
| GPQA diamond | Knowledge | 65.8% | 39.0 | — | — | Epoch AI |
| Graduate-Level Google-Proof Q&A | Knowledge | 64.2% | 37.5 | — | — | David Rein et al. |
| τ²-bench Retail | Agentic | 61.4% | 42.9 | Sierra | 30 Apr 2026 | Sierra Research |
| Artificial Analysis MMMU-Pro | Multimodal | 58.7% | 38.8 | — | — | Artificial Analysis |
| LiveCodeBench | Coding | 58.2% | 37.2 | — | 1 Sept 2026 | Vals AI |
| τ²-Bench Tool-Agent-User Evaluation | Agentic | 52.9% | 36.8 | — | — | Victor Barres et al. |
| AIME | Math | 49.4% | 36.7 | — | 16 Apr 2026 | Vals AI |
| τ²-bench Telecom | Agentic | 48.9% | 33.9 | Sierra | 2 Mar 2026 | Sierra Research |
| τ²-bench Airline | Agentic | 48.7% | 33.8 | Sierra | 2 Mar 2026 | Sierra Research |
| OTIS Mock AIME 2024-2025 | Math | 44.7% | 36.6 | — | — | Epoch AI |
| Artificial Analysis Long Context Reasoning | Reasoning | 44.0% | 38.7 | — | — | Artificial Analysis |
| Artificial Analysis IFBench | Instruction | 38.3% | 28.0 | — | — | Artificial Analysis |
| SWE-bench Verified | Coding | 23.9% | 16.7 | mini-SWE-agent | 26 Feb 2026 | SWE-bench team |
| SWE-bench Verified | Coding | 23.9% | 16.7 | mini-SWE-agent | 1 Sept 2026 | SWE-bench team |
| Software Engineering Benchmark Verified | Coding | 23.6% | 16.4 | — | — | Carlos E. Jimenez et al. |
| Artificial Analysis Omniscience Accuracy | Knowledge | 20.3% | 39.0 | — | — | Artificial Analysis |
| Artificial Analysis Coding Index | Coding | 20.2% | 33.2 | — | — | Artificial Analysis |
| SimpleQA Verified | Knowledge | 12.7% | 33.3 | — | — | Epoch AI |
| Artificial Analysis Intelligence Index | Knowledge | 10.2% | 35.1 | — | — | Artificial Analysis |
| Mystery Game Puzzles | Reasoning | 7.0% | 41.4 | — | — | Epoch AI |
| Chess Puzzles | Reasoning | 7.0% | 32.6 | — | — | Epoch AI |
| FrontierMath-Tiers-1-3-v2-Private | Math | 6.7% | 35.2 | — | — | Epoch AI |
| Artificial Analysis Humanity's Last Exam | Knowledge | 5.0% | 30.0 | — | — | Artificial Analysis |
| FrontierMath-2025-02-28-Private | Math | 4.5% | 34.8 | — | — | Epoch AI |
| ARC-AGI-1 (semi-private) | Reasoning | 3.5% | 28.9 | — | — | ARC Prize Foundation |
| GDPval-AA normalized | Agentic | 0.0% | 35.6 | — | — | Artificial Analysis |
| Critical Physics Tasks | Reasoning | 0.0% | 38.4 | — | — | Artificial Analysis |
| ARC-AGI-2 (semi-private) | Reasoning | 0.0% | 39.7 | — | — | ARC Prize Foundation |
28 benchmarks count, from 35 of 37 results. A grey row does not count. Too few models took that benchmark.