OpenAI
availableShows if the model has enough results for an index.GPT-4.1
GPT-4.1 is a non-reasoning model from OpenAI. 30 benchmarks count toward its score, in 8 categories.
IndexOverall score out of 100.38.5 ±3.0
CoverageShare of the index weight with results.100%
SpeedOutput tokens per second.53/s
Input / 1MUS dollars per 1M input tokens.$2 batch $1
Output / 1MUS dollars per 1M output tokens.$8 batch $4 US dollars per 1M output tokens in a batch.
ContextMaximum tokens in one request.1.05M
EloLMArena rating and rank.1383 (#157)
The index is a score out of 100. The ± range shows how much it can change. Batch work costs less.
49,941 votes. Elo shows what people prefer. It does not change the score.
CapabilitiesScore per category, out of 100.
Out of 100Results
30 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. |
|---|---|---|---|---|---|---|
| Massive Multitask Language Understanding | Knowledge | 90.2% | — | — | — | Dan Hendrycks et al. |
| MGSM | Multilingual | 87.7% | — | — | 9 Jan 2026 | Vals AI |
| Instruction-Following Eval | Instruction | 87.4% | 39.6 | — | — | Jeffrey Zhou et al. |
| MATH 500 | Math | 87.2% | 39.3 | — | 9 Jan 2026 | Vals AI |
| MATH level 5 | Math | 83.0% | 39.5 | — | — | Epoch AI |
| MMLU Pro | Knowledge | 80.5% | 47.3 | — | 1 Sept 2026 | Vals AI |
| τ²-bench Retail | Agentic | 74.0% | 51.9 | Sierra | 30 Apr 2026 | Sierra Research |
| MMMU Pro | Multimodal | 72.4% | 45.1 | — | 1 Sept 2026 | Vals AI |
| Artificial Analysis Long Context Reasoning | Reasoning | 68.3% | 55.5 | — | — | Artificial Analysis |
| GPQA diamond | Knowledge | 66.9% | 40.0 | — | — | Epoch AI |
| Artificial Analysis GPQA Diamond | Knowledge | 66.6% | 36.8 | — | — | Artificial Analysis |
| Graduate-Level Google-Proof Q&A | Knowledge | 66.3% | 39.5 | — | — | David Rein et al. |
| GPQA Diamond | Knowledge | 65.4% | 38.6 | — | 1 Sept 2026 | Vals AI |
| EuroEval Swedish | Multilingual | 63.8% | 82.0 | — | — | EuroEval |
| EuroEval Dutch | Multilingual | 62.6% | 80.5 | — | — | EuroEval |
| EuroEval French | Multilingual | 62.3% | 80.1 | — | — | EuroEval |
| Artificial Analysis MMMU-Pro | Multimodal | 61.2% | 41.9 | — | — | Artificial Analysis |
| EuroEval Italian | Multilingual | 57.1% | 73.6 | — | — | EuroEval |
| τ²-bench Airline | Agentic | 56.0% | 39.0 | Sierra | 2 Mar 2026 | Sierra Research |
| LiveCodeBench | Coding | 54.7% | 34.0 | — | 1 Sept 2026 | Vals AI |
| Software Engineering Benchmark Verified | Coding | 54.6% | 41.3 | — | — | Carlos E. Jimenez et al. |
| EuroEval Portuguese | Multilingual | 54.4% | 70.3 | — | — | EuroEval |
| EuroEval Polish | Multilingual | 54.1% | 69.9 | — | — | EuroEval |
| EuroEval Spanish | Multilingual | 52.8% | 68.2 | — | — | EuroEval |
| EuroEval German | Multilingual | 51.9% | 67.1 | — | — | EuroEval |
| SWE-Bench verified | Coding | 48.5% | 36.5 | — | — | Epoch AI |
| τ²-Bench Tool-Agent-User Evaluation | Agentic | 47.1% | 32.6 | — | — | Victor Barres et al. |
| Artificial Analysis IFBench | Instruction | 43.0% | 32.7 | — | — | Artificial Analysis |
| AIME | Math | 39.6% | 32.2 | — | 16 Apr 2026 | Vals AI |
| SWE-bench Verified | Coding | 39.6% | 29.3 | mini-SWE-agent | 26 Feb 2026 | SWE-bench team |
| SWE-bench Verified | Coding | 39.6% | 29.3 | mini-SWE-agent | 1 Sept 2026 | SWE-bench team |
| OTIS Mock AIME 2024-2025 | Math | 38.3% | 33.0 | — | — | Epoch AI |
| τ²-bench Telecom | Agentic | 34.0% | 23.2 | Sierra | 2 Mar 2026 | Sierra Research |
| Terminal-Bench 1.0 | Agentic | 33.8% | 40.2 | — | 12 Jan 2026 | Vals AI |
| SWE-bench Multimodal | Multimodal | 31.1% | — | GUIRepair | 17 Nov 2025 | SWE-bench team |
| SWE-bench Multimodal | Multimodal | 31.1% | — | GUIRepair | 17 Nov 2025 | SWE-bench team |
| SimpleQA Verified | Knowledge | 31.1% | 50.3 | — | — | Epoch AI |
| Artificial Analysis Omniscience Accuracy | Knowledge | 27.8% | 48.3 | — | — | Artificial Analysis |
| Gert Labs Composite Game Benchmark | Agentic | 25.6% | 35.4 | — | — | Gert Labs |
| Terminal-Bench 2.0 | Agentic | 14.6% | 33.1 | — | 4 Jun 2026 | Vals AI |
| Artificial Analysis Intelligence Index | Knowledge | 12.7% | 38.3 | — | — | Artificial Analysis |
| Chess Puzzles | Reasoning | 6.0% | 31.3 | — | — | Epoch AI |
| FrontierMath-Tiers-1-3-v2-Private | Math | 6.0% | 34.8 | — | — | Epoch AI |
| FrontierMath-2025-02-28-Private | Math | 5.5% | 35.8 | — | — | Epoch AI |
| ARC-AGI-1 (semi-private) | Reasoning | 5.5% | 29.8 | — | — | ARC Prize Foundation |
| Artificial Analysis Humanity's Last Exam | Knowledge | 4.2% | 29.1 | — | — | Artificial Analysis |
| ARC-AGI-2 (semi-private) | Reasoning | 0.4% | 39.9 | — | — | ARC Prize Foundation |
| Critical Physics Tasks | Reasoning | 0.0% | 38.4 | — | — | Artificial Analysis |
| FrontierMath-Tier-4-2025-07-01-Private | Math | 0.0% | 43.4 | — | — | Epoch AI |
30 benchmarks count, from 45 of 49 results. A grey row does not count. Too few models took that benchmark.