BenchAtlas

Rankings / Mistral AI

Mistral Large 3

open weights

released 2025-12-02

BenchAtlas Index

as of 2026-09-11
25.0
base configuration
rank #406
9 families · 5 categories · high

Benchmark evidence

42 results
agentic coding
τ²-Bench24.6 %independentArtificial Analysis
τ²-Bench
subset=banking
5.8 %independentArtificial Analysis
coding
LiveCodeBench v6
implementation=artificial-analysis
46.5 %independentArtificial Analysis
SciCode36.6 %independentArtificial Analysis
external indices
AA Coding Index20.2 pointsindependentArtificial Analysis
Intelligence Index v4.111.1 pointsindependentArtificial Analysis
AA Math Index38.0 pointsindependentArtificial Analysis
human preference
Coding (style control)older version
arena=text · category=coding · style_control=true
1469.0
1463.8–1474.2
communityLMArena Leaderboard Dataset
Industry Software And It Services (style control)older version
arena=text · category=industry_software_and_it_services · style_control=true
1456.2
1451.7–1460.7
communityLMArena Leaderboard Dataset
Hard Prompts English (style control)older version
arena=text · category=hard_prompts_english · style_control=true
1438.7
1433.4–1443.9
communityLMArena Leaderboard Dataset
French (style control)older version
arena=text · category=french · style_control=true
1435.2
1419.7–1450.6
communityLMArena Leaderboard Dataset
Chinese (style control)older version
arena=text · category=chinese · style_control=true
1431.4
1422.2–1440.7
communityLMArena Leaderboard Dataset
Industry Medicine And Healthcare (style control)older version
arena=text · category=industry_medicine_and_healthcare · style_control=true
1431.4
1422.0–1440.8
communityLMArena Leaderboard Dataset
Hard Prompts (style control)older version
arena=text · category=hard_prompts · style_control=true
1431.3
1427.4–1435.1
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
1427.7
1421.6–1433.9
communityLMArena Leaderboard Dataset
English (style control)older version
arena=text · category=english · style_control=true
1427.1
1422.8–1431.3
communityLMArena Leaderboard Dataset
Expert (style control)older version
arena=text · category=expert · style_control=true
1423.9
1415.6–1432.2
communityLMArena Leaderboard Dataset
Multi Turn (style control)older version
arena=text · category=multi_turn · style_control=true
1421.0
1414.9–1427.2
communityLMArena Leaderboard Dataset
Industry Legal And Government (style control)older version
arena=text · category=industry_legal_and_government · style_control=true
1420.4
1411.5–1429.3
communityLMArena Leaderboard Dataset
Industry Mathematical (style control)older version
arena=text · category=industry_mathematical · style_control=true
1418.1
1407.1–1429.0
communityLMArena Leaderboard Dataset
Longer Query (style control)older version
arena=text · category=longer_query · style_control=true
1417.3
1412.5–1422.0
communityLMArena Leaderboard Dataset
Polish (style control)older version
arena=text · category=polish · style_control=true
1415.7
1400.6–1430.7
communityLMArena Leaderboard Dataset
Overall (style control)
arena=text · category=overall · style_control=true
1413.7
1410.6–1416.7
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
1412.8
1406.9–1418.7
communityLMArena Leaderboard Dataset
Russian (style control)older version
arena=text · category=russian · style_control=true
1410.1
1403.0–1417.2
communityLMArena Leaderboard Dataset
Spanish (style control)older version
arena=text · category=spanish · style_control=true
1407.0
1392.6–1421.3
communityLMArena Leaderboard Dataset
German (style control)older version
arena=text · category=german · style_control=true
1403.7
1384.7–1422.7
communityLMArena Leaderboard Dataset
Math (style control)older version
arena=text · category=math · style_control=true
1402.5
1392.6–1412.4
communityLMArena Leaderboard Dataset
Exclude Ties (style control)older version
arena=text · category=exclude_ties · style_control=true
1401.1
1396.8–1405.4
communityLMArena Leaderboard Dataset
Instruction Following (style control)older version
arena=text · category=instruction_following · style_control=true
1400.3
1395.5–1405.1
communityLMArena Leaderboard Dataset
Non English (style control)older version
arena=text · category=non_english · style_control=true
1398.0
1394.2–1401.8
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
1390.6
1385.2–1395.9
communityLMArena Leaderboard Dataset
Japanese (style control)older version
arena=text · category=japanese · style_control=true
1383.0
1358.6–1407.4
communityLMArena Leaderboard Dataset
Creative Writing (style control)older version
arena=text · category=creative_writing · style_control=true
1375.0
1368.7–1381.3
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
1373.3
1367.7–1379.0
communityLMArena Leaderboard Dataset
Korean (style control)older version
arena=text · category=korean · style_control=true
1355.8
1336.7–1374.9
communityLMArena Leaderboard Dataset
knowledge science
GPQA Diamond
implementation=artificial-analysis
68.0 %independentArtificial Analysis
Humanity's Last Exam
implementation=artificial-analysis
4.2 %independentArtificial Analysis
MMLU-Pro
implementation=artificial-analysis
80.7 %independentArtificial Analysis
long context instruction
AA-LCR36.0 %independentArtificial Analysis
IFBench36.2 %independentArtificial Analysis
reasoning math
AIME
year=2025 · implementation=artificial-analysis
38.0 %independentArtificial Analysis

Agent + model results

systems, not bare-model scores
agent + model Artificial Analysis harness + Mistral Large 3Terminal-Bench Hard15.9 %independentArtificial Analysis
agent + model Artificial Analysis harness + Mistral Large 3Terminal-Bench 2.112.0 %independentArtificial Analysis

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

Related variants

Mistral Large 2411Mistral Large 2402Mistral Large 2407Mistral Large 2512Mistral Large