Rankings / OpenAI
GPT 4.1 Mini (2025-04-14)
released 2025-04-14
BenchAtlas Index
as of 2026-07-14
high effort
rank #370
8 families · 4 categories · medium
base configuration
rank #376
15 families · 6 categories · high
Benchmark evidence
68 results
agentic coding
| τ²-Bench | 52.9 % | independent | Artificial Analysis | |
| τ²-Bench subset=banking | 5.2 % | independent | Artificial Analysis |
coding
| LiveCodeBench v6 implementation=artificial-analysis | 48.3 % | independent | Artificial Analysis | |
| SciCode | 40.4 % | independent | Artificial Analysis |
external indices
| AA Coding Index | 20.2 points | independent | Artificial Analysis | |
| Intelligence Index v4.1 | 14.8 points | independent | Artificial Analysis | |
| AA Math Index | 46.3 points | independent | Artificial Analysis | |
| ECI | 135.8 points | independent | Epoch AI Benchmarking Hub |
factuality
| MASK contamination=Potential contamination warning: This model was evaluated after the public release of MASK, allowing model builder access to the prompts and solutions. | 50.0 % 47.8–52.2 | independent | Scale Labs |
human preference
| Coding (style control)older version arena=text · category=coding · style_control=true | 1433.3 1425.8–1440.8 | community | LMArena Leaderboard Dataset | |
| Industry Software And It Services (style control)older version arena=text · category=industry_software_and_it_services · style_control=true | 1418.7 1412.7–1424.8 | community | LMArena Leaderboard Dataset | |
| Hard Prompts English (style control)older version arena=text · category=hard_prompts_english · style_control=true | 1413.0 1406.1–1420.0 | community | LMArena Leaderboard Dataset | |
| French (style control)older version arena=text · category=french · style_control=true | 1405.8 1379.0–1432.5 | community | LMArena Leaderboard Dataset | |
| Hard Prompts (style control)older version arena=text · category=hard_prompts · style_control=true | 1402.3 1396.7–1408.0 | community | LMArena Leaderboard Dataset | |
| Industry Legal And Government (style control)older version arena=text · category=industry_legal_and_government · style_control=true | 1398.2 1386.2–1410.2 | community | LMArena Leaderboard Dataset | |
| English (style control)older version arena=text · category=english · style_control=true | 1397.9 1392.6–1403.2 | community | LMArena Leaderboard Dataset | |
| Industry Medicine And Healthcare (style control)older version arena=text · category=industry_medicine_and_healthcare · style_control=true | 1397.7 1385.0–1410.4 | community | LMArena Leaderboard Dataset | |
| Multi Turn (style control)older version arena=text · category=multi_turn · style_control=true | 1392.1 1384.5–1399.7 | community | LMArena Leaderboard Dataset | |
| German (style control)older version arena=text · category=german · style_control=true | 1391.0 1372.2–1409.9 | community | LMArena Leaderboard Dataset | |
| Longer Query (style control)older version arena=text · category=longer_query · style_control=true | 1389.2 1381.7–1396.7 | community | LMArena Leaderboard Dataset | |
| Industry Life And Physical And Social Science (style control)older version arena=text · category=industry_life_and_physical_and_social_science · style_control=true | 1384.4 1376.5–1392.3 | community | LMArena Leaderboard Dataset | |
| Chinese (style control)older version arena=text · category=chinese · style_control=true | 1383.7 1370.3–1397.0 | community | LMArena Leaderboard Dataset | |
| Expert (style control)older version arena=text · category=expert · style_control=true | 1382.9 1369.8–1395.9 | community | LMArena Leaderboard Dataset | |
| Overall (style control) arena=text · category=overall · style_control=true | 1382.8 1378.5–1387.1 | community | LMArena Leaderboard Dataset | |
| Polish (style control)older version arena=text · category=polish · style_control=true | 1382.4 1370.0–1394.9 | community | LMArena Leaderboard Dataset | |
| Industry Business And Management And Financial Operations (style control)older version arena=text · category=industry_business_and_management_and_financial_operations · style_control=true | 1379.8 1371.7–1387.9 | community | LMArena Leaderboard Dataset | |
| Instruction Following (style control)older version arena=text · category=instruction_following · style_control=true | 1372.7 1366.2–1379.1 | community | LMArena Leaderboard Dataset | |
| Industry Mathematical (style control)older version arena=text · category=industry_mathematical · style_control=true | 1364.5 1352.9–1376.1 | community | LMArena Leaderboard Dataset | |
| Russian (style control)older version arena=text · category=russian · style_control=true | 1361.0 1349.7–1372.4 | community | LMArena Leaderboard Dataset | |
| Industry Writing And Literature And Language (style control)older version arena=text · category=industry_writing_and_literature_and_language · style_control=true | 1360.2 1353.3–1367.2 | community | LMArena Leaderboard Dataset | |
| Non English (style control)older version arena=text · category=non_english · style_control=true | 1359.7 1354.3–1365.0 | community | LMArena Leaderboard Dataset | |
| Exclude Ties (style control)older version arena=text · category=exclude_ties · style_control=true | 1359.0 1353.2–1364.9 | community | LMArena Leaderboard Dataset | |
| Spanish (style control)older version arena=text · category=spanish · style_control=true | 1357.1 1334.6–1379.6 | community | LMArena Leaderboard Dataset | |
| Math (style control)older version arena=text · category=math · style_control=true | 1354.0 1342.9–1365.1 | community | LMArena Leaderboard Dataset | |
| Creative Writing (style control)older version arena=text · category=creative_writing · style_control=true | 1349.4 1340.7–1358.2 | community | LMArena Leaderboard Dataset | |
| Industry Entertainment And Sports And Media (style control)older version arena=text · category=industry_entertainment_and_sports_and_media · style_control=true | 1348.5 1340.9–1356.1 | community | LMArena Leaderboard Dataset | |
| Korean (style control)older version arena=text · category=korean · style_control=true | 1347.2 1325.0–1369.4 | community | LMArena Leaderboard Dataset | |
| Japanese (style control)older version arena=text · category=japanese · style_control=true | 1324.2 1303.6–1344.7 | community | LMArena Leaderboard Dataset |
knowledge science
| GPQA Diamond implementation=vals-ai | high effort | 67.9 % | independent | Vals AI |
| GPQA Diamond implementation=artificial-analysis | 66.4 % | independent | Artificial Analysis | |
| GPQA Diamond | 65.8 % | independent | Epoch AI Benchmarking Hub 2025-04-14 | |
| Humanity's Last Exam implementation=artificial-analysis | 4.6 % | independent | Artificial Analysis | |
| MMLU-Pro implementation=artificial-analysis | 78.1 % | independent | Artificial Analysis | |
| MMLU-Pro implementation=vals-ai | high effort | 77.2 % | independent | Vals AI |
long context instruction
| AA-LCR | 42.3 % | independent | Artificial Analysis | |
| IFBench | 38.3 % | independent | Artificial Analysis |
multimodal
| MMMU implementation=vals-ai | high effort | 70.5 % | independent | Vals AI |
| VISTA | 41.1 % 40.6–41.7 | independent | Scale Labs |
professional
| CorpFin | high effort | 57.9 % | independent | Vals AI |
| LegalBench | high effort | 78.0 % | independent | Vals AI |
| MedQA | high effort | 84.6 % | independent | Vals AI |
| TaxEval | high effort | 71.9 % | independent | Vals AI |
reasoning math
| AIME implementation=vals-ai | high effort | 49.4 % | independent | Vals AI |
| AIME year=2025 · implementation=artificial-analysis | 46.3 % | independent | Artificial Analysis | |
| AIME implementation=artificial-analysis | 43.0 % | independent | Artificial Analysis | |
| ARC-AGI-1older version split=public_eval · model_type=Base LLM | 7.3 % | independent | ARC Prize Leaderboard | |
| ARC-AGI-1older version split=semi_private · model_type=Base LLM | 3.5 % | independent | ARC Prize Leaderboard | |
| ARC-AGI-2 split=semi_private · model_type=Base LLM | 0.0 usd_per_task | independent | ARC Prize Leaderboard | |
| ARC-AGI-2 split=public_eval · model_type=Base LLM | 0.0 usd_per_task | independent | ARC Prize Leaderboard | |
| ARC-AGI-1older version split=semi_private · model_type=Base LLM | 0.0 usd_per_task | independent | ARC Prize Leaderboard | |
| ARC-AGI-1older version split=public_eval · model_type=Base LLM | 0.0 usd_per_task | independent | ARC Prize Leaderboard | |
| ARC-AGI-2 split=public_eval · model_type=Base LLM | 0.0 % | independent | ARC Prize Leaderboard | |
| ARC-AGI-2 split=semi_private · model_type=Base LLM | 0.0 % | independent | ARC Prize Leaderboard | |
| FrontierMath | 4.5 % | independent | Epoch AI Benchmarking Hub 2025-04-14 | |
| MATH-500 implementation=artificial-analysis | 92.5 % | independent | Artificial Analysis | |
| MATH-500 implementation=vals-ai | high effort | 88.0 % | independent | Vals AI |
| MATH Level 5 | 87.3 % | independent | Epoch AI Benchmarking Hub 2025-04-14 | |
| OTIS Mock AIME 2024–2025 | 44.7 % | independent | Epoch AI Benchmarking Hub 2025-04-14 |
Agent + model results
systems, not bare-model scores
| agent + model Artificial Analysis harness + GPT 4.1 Mini (2025-04-14) | Terminal-Bench 2.1 | 10.1 % | independent | Artificial Analysis |
| agent + model Artificial Analysis harness + GPT 4.1 Mini (2025-04-14) | Terminal-Bench Hard | 7.6 % | independent | Artificial Analysis |
These scores measure the whole agent system (scaffold, tools, budgets) — they are never merged into the bare model’s numbers.