Rankings / MiniMax
Minimax M2 5
released 2026-02-12
BenchAtlas Index
as of 2026-09-11
base configuration
rank #127
9 families · 5 categories · high
base configuration
rank #163
9 families · 3 categories · medium
Benchmark evidence
52 results
agentic coding
| τ²-Bench | 95.3 % | independent | Artificial Analysis |
coding
| SciCode | 42.6 % | independent | Artificial Analysis |
external indices
| Intelligence Index v4.1 | 22.8 points | independent | Artificial Analysis | |
| ECI | 146.5 points | independent | Epoch AI Benchmarking Hub |
human preference
| Coding (style control)older version arena=text · category=coding · style_control=true | 1444.3 1437.7–1450.9 | community | LMArena Leaderboard Dataset | |
| Industry Software And It Services (style control)older version arena=text · category=industry_software_and_it_services · style_control=true | 1435.7 1430.0–1441.4 | community | LMArena Leaderboard Dataset | |
| Chinese (style control)older version arena=text · category=chinese · style_control=true | 1431.9 1419.5–1444.4 | community | LMArena Leaderboard Dataset | |
| Hard Prompts English (style control)older version arena=text · category=hard_prompts_english · style_control=true | 1429.5 1423.1–1436.0 | community | LMArena Leaderboard Dataset | |
| Expert (style control)older version arena=text · category=expert · style_control=true | 1420.5 1410.0–1431.0 | community | LMArena Leaderboard Dataset | |
| Hard Prompts (style control)older version arena=text · category=hard_prompts · style_control=true | 1415.5 1410.6–1420.4 | community | LMArena Leaderboard Dataset | |
| English (style control)older version arena=text · category=english · style_control=true | 1412.3 1407.0–1417.7 | 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 | 1408.9 1401.0–1416.8 | community | LMArena Leaderboard Dataset | |
| Industry Medicine And Healthcare (style control)older version arena=text · category=industry_medicine_and_healthcare · style_control=true | 1408.4 1396.7–1420.0 | community | LMArena Leaderboard Dataset | |
| Longer Query (style control)older version arena=text · category=longer_query · style_control=true | 1404.1 1398.1–1410.0 | community | LMArena Leaderboard Dataset | |
| Industry Legal And Government (style control)older version arena=text · category=industry_legal_and_government · style_control=true | 1404.1 1392.6–1415.5 | community | LMArena Leaderboard Dataset | |
| French (style control)older version arena=text · category=french · style_control=true | 1402.1 1383.3–1420.9 | community | LMArena Leaderboard Dataset | |
| Industry Mathematical (style control)older version arena=text · category=industry_mathematical · style_control=true | 1400.6 1387.8–1413.5 | 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 | 1399.8 1392.5–1407.2 | community | LMArena Leaderboard Dataset | |
| Math (style control)older version arena=text · category=math · style_control=true | 1395.6 1383.5–1407.7 | community | LMArena Leaderboard Dataset | |
| Multi Turn (style control)older version arena=text · category=multi_turn · style_control=true | 1395.3 1387.7–1402.9 | community | LMArena Leaderboard Dataset | |
| Russian (style control)older version arena=text · category=russian · style_control=true | 1392.0 1383.0–1400.9 | community | LMArena Leaderboard Dataset | |
| German (style control)older version arena=text · category=german · style_control=true | 1391.2 1366.7–1415.8 | community | LMArena Leaderboard Dataset | |
| Overall (style control) arena=text · category=overall · style_control=true | 1390.5 1386.6–1394.5 | community | LMArena Leaderboard Dataset | |
| Instruction Following (style control)older version arena=text · category=instruction_following · style_control=true | 1380.7 1374.7–1386.7 | community | LMArena Leaderboard Dataset | |
| Spanish (style control)older version arena=text · category=spanish · style_control=true | 1375.3 1358.3–1392.2 | 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 | 1371.5 1364.7–1378.2 | community | LMArena Leaderboard Dataset | |
| Exclude Ties (style control)older version arena=text · category=exclude_ties · style_control=true | 1371.4 1366.0–1376.8 | community | LMArena Leaderboard Dataset | |
| Polish (style control)older version arena=text · category=polish · style_control=true | 1370.5 1350.5–1390.5 | community | LMArena Leaderboard Dataset | |
| Non English (style control)older version arena=text · category=non_english · style_control=true | 1367.3 1362.3–1372.2 | 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 | 1362.7 1355.4–1370.0 | community | LMArena Leaderboard Dataset | |
| Creative Writing (style control)older version arena=text · category=creative_writing · style_control=true | 1358.6 1350.5–1366.7 | community | LMArena Leaderboard Dataset | |
| Korean (style control)older version arena=text · category=korean · style_control=true | 1270.7 1245.7–1295.8 | community | LMArena Leaderboard Dataset |
knowledge science
| GPQA Diamond implementation=artificial-analysis | 84.8 % | independent | Artificial Analysis | |
| GPQA Diamond implementation=vals-ai | 82.1 % | independent | Vals AI | |
| Humanity's Last Exam implementation=artificial-analysis | 20.5 % | independent | Artificial Analysis | |
| MMLU-Pro implementation=vals-ai | 80.1 % | independent | Vals AI |
long context instruction
| AA-LCR | 73.3 % | independent | Artificial Analysis | |
| IFBench | 71.6 % | independent | Artificial Analysis |
professional
| CaseLaw | 53.5 % | independent | Vals AI | |
| CorpFin | 59.6 % | independent | Vals AI | |
| LegalBench | 80.0 % | independent | Vals AI | |
| MedQA | 92.5 % | independent | Vals AI | |
| TaxEval | 68.2 % | independent | Vals AI |
reasoning math
| AIME implementation=vals-ai | 88.8 % | independent | Vals AI | |
| ARC-AGI-1older version split=semi_private · model_type=CoT | 63.7 % | independent | ARC Prize Leaderboard | |
| ARC-AGI-1older version split=public_eval · model_type=CoT | 59.1 % | independent | ARC Prize Leaderboard | |
| ARC-AGI-2 split=public_eval · model_type=CoT | 5.4 % | independent | ARC Prize Leaderboard | |
| ARC-AGI-2 split=semi_private · model_type=CoT | 4.9 % | independent | ARC Prize Leaderboard | |
| ARC-AGI-2 split=public_eval · model_type=CoT | 0.2 usd_per_task | independent | ARC Prize Leaderboard | |
| ARC-AGI-2 split=semi_private · model_type=CoT | 0.2 usd_per_task | independent | ARC Prize Leaderboard | |
| ARC-AGI-1older version split=public_eval · model_type=CoT | 0.1 usd_per_task | independent | ARC Prize Leaderboard | |
| ARC-AGI-1older version split=semi_private · model_type=CoT | 0.1 usd_per_task | independent | ARC Prize Leaderboard |
Agent + model results
systems, not bare-model scores
| agent + model mini-SWE-agent + Minimax M2 5 | SWE-bench bash-only | 75.8 % | unverified | SWE-bench Leaderboard |
| agent + model mini-SWE-agent + Minimax M2 5 | SWE-bench Verified | 75.8 % | unverified | SWE-bench Leaderboard |
| agent + model mini-SWE-agent + Minimax M2 5 | SWE-bench Verified | 0.1 usd_per_task | unverified | SWE-bench Leaderboard |
| agent + model mini-SWE-agent + Minimax M2 5 | SWE-bench bash-only | 0.1 usd_per_task | unverified | SWE-bench Leaderboard |
| agent + model cchuter + Minimax M2 5 | Terminal-Bench 2.0 | 42.7 % | unverified | Terminal-Bench Leaderboard |
| agent + model terminus-2 + Minimax M2 5 | Terminal-Bench 2.0 | 42.3 % | unverified | Terminal-Bench Leaderboard |
| agent + model Artificial Analysis harness + Minimax M2 5 | Terminal-Bench Hard | 34.9 % | independent | Artificial Analysis |
These scores measure the whole agent system (scaffold, tools, budgets) — they are never merged into the bare model’s numbers.