BenchAtlas

Rankings / MiniMax

Minimax M3

released 2026-06-01

BenchAtlas Index

as of 2026-09-11
64.5
base configuration
rank #116
7 families · 4 categories · medium
42.8
base configuration
rank #290
13 families · 5 categories · high

Benchmark evidence

58 results
agentic coding
Release 2026-06-25
tasks_counted=3 · livebench_version=2026-06-25
40.7 %independentLiveBench
τ²-Bench88.9 %independentArtificial Analysis
τ²-Bench
subset=banking
15.3 %independentArtificial Analysis
coding
Release 2026-06-25
tasks_counted=2 · livebench_version=2026-06-25
68.2 %independentLiveBench
SciCode47.1 %independentArtificial Analysis
data analysis
Release 2026-06-25
tasks_counted=3 · livebench_version=2026-06-25
76.2 %independentLiveBench
external indices
AA Coding Index58.6 pointsindependentArtificial Analysis
Intelligence Index v4.129.6 pointsindependentArtificial Analysis
ECI146.6 pointsindependentEpoch AI Benchmarking Hub
ECIno reasoning146.6 pointsindependentEpoch AI Benchmarking Hub
Vals Index42.7 pointsindependentVals AI
human preference
Coding (style control)older version
arena=text · category=coding · style_control=true
1498.0
1491.5–1504.4
communityLMArena Leaderboard Dataset
Industry Software And It Services (style control)older version
arena=text · category=industry_software_and_it_services · style_control=true
1487.2
1481.4–1493.0
communityLMArena Leaderboard Dataset
Hard Prompts English (style control)older version
arena=text · category=hard_prompts_english · style_control=true
1475.0
1468.6–1481.5
communityLMArena Leaderboard Dataset
Expert (style control)older version
arena=text · category=expert · style_control=true
1474.5
1465.1–1483.8
communityLMArena Leaderboard Dataset
Chinese (style control)older version
arena=text · category=chinese · style_control=true
1469.9
1457.0–1482.9
communityLMArena 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
1466.0
1458.2–1473.8
communityLMArena Leaderboard Dataset
Industry Medicine And Healthcare (style control)older version
arena=text · category=industry_medicine_and_healthcare · style_control=true
1465.0
1453.4–1476.6
communityLMArena Leaderboard Dataset
Hard Prompts (style control)older version
arena=text · category=hard_prompts · style_control=true
1464.0
1458.9–1469.0
communityLMArena Leaderboard Dataset
English (style control)older version
arena=text · category=english · style_control=true
1455.6
1449.9–1461.2
communityLMArena Leaderboard Dataset
Longer Query (style control)older version
arena=text · category=longer_query · style_control=true
1453.5
1447.6–1459.3
communityLMArena Leaderboard Dataset
French (style control)older version
arena=text · category=french · style_control=true
1452.8
1436.3–1469.3
communityLMArena Leaderboard Dataset
Multi Turn (style control)older version
arena=text · category=multi_turn · style_control=true
1451.2
1443.5–1458.9
communityLMArena Leaderboard Dataset
Industry Mathematical (style control)older version
arena=text · category=industry_mathematical · style_control=true
1451.0
1438.2–1463.9
communityLMArena 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
1450.7
1443.3–1458.0
communityLMArena Leaderboard Dataset
Russian (style control)older version
arena=text · category=russian · style_control=true
1443.2
1434.0–1452.4
communityLMArena Leaderboard Dataset
Overall (style control)
arena=text · category=overall · style_control=true
1443.0
1438.6–1447.3
communityLMArena Leaderboard Dataset
Exclude Ties (style control)older version
arena=text · category=exclude_ties · style_control=true
1440.7
1435.0–1446.4
communityLMArena Leaderboard Dataset
Math (style control)older version
arena=text · category=math · style_control=true
1438.0
1424.7–1451.3
communityLMArena Leaderboard Dataset
German (style control)older version
arena=text · category=german · style_control=true
1437.3
1415.3–1459.4
communityLMArena Leaderboard Dataset
Industry Legal And Government (style control)older version
arena=text · category=industry_legal_and_government · style_control=true
1436.3
1425.7–1447.0
communityLMArena Leaderboard Dataset
Instruction Following (style control)older version
arena=text · category=instruction_following · style_control=true
1435.0
1429.0–1441.1
communityLMArena Leaderboard Dataset
Polish (style control)older version
arena=text · category=polish · style_control=true
1434.7
1414.4–1455.1
communityLMArena Leaderboard Dataset
Spanish (style control)older version
arena=text · category=spanish · style_control=true
1427.6
1410.2–1445.0
communityLMArena Leaderboard Dataset
Non English (style control)older version
arena=text · category=non_english · style_control=true
1427.3
1422.1–1432.5
communityLMArena Leaderboard Dataset
Industry Writing And Literature And Language (style control)older version
arena=text · category=industry_writing_and_literature_and_language · style_control=true
1422.9
1416.1–1429.7
communityLMArena Leaderboard Dataset
Creative Writing (style control)older version
arena=text · category=creative_writing · style_control=true
1407.1
1399.3–1414.9
communityLMArena Leaderboard Dataset
Industry Entertainment And Sports And Media (style control)older version
arena=text · category=industry_entertainment_and_sports_and_media · style_control=true
1406.8
1399.8–1413.9
communityLMArena Leaderboard Dataset
Japanese (style control)older version
arena=text · category=japanese · style_control=true
1398.3
1372.2–1424.4
communityLMArena Leaderboard Dataset
Korean (style control)older version
arena=text · category=korean · style_control=true
1370.0
1348.3–1391.6
communityLMArena Leaderboard Dataset
knowledge science
GPQA Diamond
implementation=artificial-analysis
92.9 %independentArtificial Analysis
GPQA Diamond
implementation=vals-ai
92.7 %independentVals AI
GPQA Diamond90.9 %independentEpoch AI Benchmarking Hub
2026-08-10
GPQA Diamondno reasoning81.3 %independentEpoch AI Benchmarking Hub
2026-08-10
Humanity's Last Exam
implementation=artificial-analysis
39.0 %independentArtificial Analysis
MMLU-Pro
implementation=vals-ai
84.2 %independentVals AI
language
Release 2026-06-25
tasks_counted=3 · livebench_version=2026-06-25
76.8 %independentLiveBench
long context instruction
AA-LCR83.0 %independentArtificial Analysis
IFBench82.9 %independentArtificial Analysis
Release 2026-06-25
tasks_counted=4 · livebench_version=2026-06-25
57.5 %independentLiveBench
multimodal
MMMU
implementation=vals-ai
81.2 %independentVals AI
professional
CorpFin68.1 %independentVals AI
LegalBench85.4 %independentVals AI
TaxEval72.7 %independentVals AI
reasoning math
Release 2026-06-25
tasks_counted=4 · livebench_version=2026-06-25
77.0 %independentLiveBench
Release 2026-06-25
tasks_counted=4 · livebench_version=2026-06-25
74.5 %independentLiveBench
OTIS Mock AIME 2024–202571.1 %independentEpoch AI Benchmarking Hub
2026-08-10
OTIS Mock AIME 2024–2025no reasoning26.7 %independentEpoch AI Benchmarking Hub
2026-08-10

Agent + model results

systems, not bare-model scores
agent + model Artificial Analysis harness + Minimax M3Terminal-Bench 2.165.2 %independentArtificial Analysis
agent + model Artificial Analysis harness + Minimax M3Terminal-Bench Hard42.4 %independentArtificial Analysis

These scores measure the whole agent system (scaffold, tools, budgets) — they are never merged into the bare model’s numbers.