BenchLeader

MiniMax M3.1 Flash vs Mistral Large 4

Full profiles: MiniMax M3.1 Flash (MiniMax) and Mistral Large 4 (Mistral AI): every benchmark, price and rank.

Updated · Data from 8 sources

As of 16 Nov 2026, MiniMax M3.1 Flash and Mistral Large 4 are neck and neck on quality (54.3 vs 52.9 on the BenchLeader Index, inside the ±6.4 margin), and MiniMax M3.1 Flash is 4.0× cheaper at $0.262 vs $1.05 per 1M tokens blended. MiniMax M3.1 Flash leads in every category we measure both in. MiniMax M3.1 Flash streams faster.

Key facts (as of 16 Nov 2026)

  • MiniMax M3.1 Flash scores 54.3 on the BenchLeader Index and Mistral Large 4 scores 52.9 (BenchLeader, 16 Nov 2026).
  • MiniMax M3.1 Flash costs $0.262 per 1M tokens blended (list price, 16 Nov 2026).
  • Mistral Large 4 costs $1.05 per 1M tokens blended (list price, 16 Nov 2026).
  • MiniMax M3.1 Flash streams faster than Mistral Large 4 (Artificial Analysis, 7-day median, 16 Nov 2026).
  • Of the 26 benchmarks both models have, MiniMax M3.1 Flash leads 19 (BenchLeader, 16 Nov 2026).
  • On Arena WebDev, MiniMax M3.1 Flash is rated 1584 and Mistral Large 4 1565 (LMArena, as published by 16 Nov 2026).

Which to pick for…

The pick's numbers come first. Category scores on the BenchLeader scale (50 is the average model, 15 points one standard deviation). High confidence means a lead of half a standard deviation or more; under one point is called either.

Use casePickWhyConfidence
CodingMiniMax M3.1 Flash55 vs 54 category score; Arena WebDev 1584 vs 1565Medium
Agents & tool useMiniMax M3.1 Flash55 vs 49 category score; Arena Agent -2.7 vs -5.4Medium
ReasoningMiniMax M3.1 Flash50 vs 46 category score; HLE-Diamond (AA) 28.1% vs 15.1%Medium
KnowledgeMiniMax M3.1 Flash46 vs 41 category score; AA-Omniscience v1.1 -6.4 vs -7.5Medium
Long documentsEither65 vs 65 category score; AA-LCR v1.1 79.7% vs 81.3%Either
Images and other mediaMiniMax M3.1 Flash62 vs 58 category score; MMMU-Pro 79.1% vs 76.4%Medium
Chat quality (human preference)–Only Mistral Large 4 has an Arena rating (60 category score)Not comparable
Lowest costMiniMax M3.1 Flash$0.262 vs $1.05 per 1M blended; agentic coding session $0.011 vs $0.050–
Fastest first answerEither43.3 s (MiniMax M3.1 Flash) vs 40.9 s (Mistral Large 4) to the first answer token–
High volume on a budgetNeither flagshipA smaller model is usually the better fit: Cheapest good LLMs, Fastest LLMs–

Key differences

  • MiniMax M3.1 Flash and Mistral Large 4 are neck and neck on quality (54.3 vs 52.9, within the ±6.4 margin).
  • MiniMax M3.1 Flash is stronger in agents & tools, coding, knowledge, multimodal, reasoning.
  • Mistral Large 4 is stronger in long context.
  • MiniMax M3.1 Flash is 4.0× cheaper ($0.262 vs $1.05 per 1M blended).
  • MiniMax M3.1 Flash streams 1.6× faster (151 vs 97 tokens per second).

At a glance

SpecMiniMax M3.1 FlashMistral Large 4
DeveloperMiniMaxMistral AI
Released19 Oct 20266 Oct 2026
Context window (tokens)1,000,0001,048,576
Compared atMiniMax M3.1 FlashMistral Large 4
Open weightsYesYes
Price in / out / cached input per 1M$0.150 / $0.600 / $0.015$0.700 / $2.10 / $0.070

Category by category

Category scores on the BenchLeader scale: 50 is the average evaluated model, 15 points is one standard deviation. A gap under one point is called level.

CategoryMiniMax M3.1 FlashMistral Large 4Leader
Reasoning50.146.2MiniMax M3.1 Flash
Coding55.453.8MiniMax M3.1 Flash
Agents & tools55.248.8MiniMax M3.1 Flash
Maths–54.2not comparable
Knowledge45.940.7MiniMax M3.1 Flash
Human preference–59.7not comparable
Multimodal61.558.0MiniMax M3.1 Flash
Long context64.565.3level

On one chart

Each domain on the index's own scale, where 50 is an average model. Click a domain to see the benchmarks behind it. Pick up to 5 models from the list.

2550 avg75ReasoningCodingAgents & toolsKnowledgeLong contextMultimodalMiniMax M3.1 Flash, Reasoning: 50.1MiniMax M3.1 Flash, Coding: 55.4MiniMax M3.1 Flash, Agents & tools: 55.2MiniMax M3.1 Flash, Knowledge: 45.9MiniMax M3.1 Flash, Long context: 64.5MiniMax M3.1 Flash, Multimodal: 61.5Mistral Large 4, Reasoning: 46.2Mistral Large 4, Coding: 53.8Mistral Large 4, Agents & tools: 48.8Mistral Large 4, Knowledge: 40.7Mistral Large 4, Long context: 65.3Mistral Large 4, Multimodal: 58.0

Not drawn, measured for fewer than two of these models: Maths, Instruction following, Human preference.

Model, click to add or removeIndex

Where sources disagree

Across the 26 benchmarks both have, MiniMax M3.1 Flash leads 19, Mistral Large 4 leads 7. Every overall index that rates both puts MiniMax M3.1 Flash ahead.

Overall indexMiniMax M3.1 FlashMistral Large 4Leader
BenchLeader Index54.352.9neck and neck
AA Intelligence Index v5.0.129.125.3MiniMax M3.1 Flash

Each publisher runs its own tasks, harnesses and effort settings, so their overall verdicts can differ; BenchLeader reconciles them on one scale. † measured at another effort setting of the same model (the publisher did not test the one compared here).

Price, speed and latency

List prices blended at three input tokens per output token. Per-task prices use the token counts on the methodology page. Cost, tokens and time per benchmark task are measured: Artificial Analysis’ averages over its whole Intelligence Index run.

MeasureMiniMax M3.1 FlashMistral Large 4
Blended price per 1M tokens$0.262$1.05
Input / output per 1M$0.150 / $0.600$0.700 / $2.10
Cached input per 1M$0.015$0.070
Price per task: chat reply$0.0002$0.0009
Price per task: summarize a 30-page report$0.0022$0.0097
Price per task: code edit$0.0018$0.0074
Price per task: agentic coding session$0.011$0.050
Price per task: structured extraction$0.0004$0.0018
Agentic coding session, input mostly cached$0.0053$0.022
Cost per benchmark task (AA Index)$0.491$4.21
Output tokens per task (AA Index)138k135k
Of which reasoning103k102k
Time per task (AA Index)15 min23 min
Output speed151 tok/s97 tok/s
Time to first answer43.3 s40.9 s
Time to a complete answer, last measured 13 Nov 2026141.5 s119.0 s

Per token is not per task: an average task of Artificial Analysis' Intelligence Index cost $0.491 on MiniMax M3.1 Flash and $4.21 on Mistral Large 4, and MiniMax M3.1 Flash wrote 1.0× as many output tokens for the same work.

Each model is compared at its best-ranked reasoning-effort setting. MiniMax M3.1 Flash scores 54.3 and takes 43 s to its first answer token. Mistral Large 4 scores 52.9 and takes 41 s to its first answer token; Mistral Large 4 (high) scores 48.3 in 45 s, Mistral Large 4 (medium) scores 45.9 in 27 s. Higher effort means more thinking before the answer: better scores, longer waits and more tokens billed. How reasoning effort changes scores and cost.

Every benchmark

123 benchmarks, grouped by category; each name links to the board and its leaders. † measured at another effort setting of the same model, shown for reference and not counted as a win.

MetricPublisherMiniMax M3.1 FlashMistral Large 4
BenchLeader IndexBenchLeader54.352.9
Agents & tools scoreBenchLeader55.248.8
Coding scoreBenchLeader55.453.8
Knowledge scoreBenchLeader45.940.7
Long context scoreBenchLeader64.565.3
Multimodal scoreBenchLeader61.558.0
Reasoning scoreBenchLeader50.146.2
Human preference scoreBenchLeader–59.7
Maths scoreBenchLeader–54.2
Blended price $/MBenchLeader$0.262$1.05
Output speedBenchLeader151 tok/s97 tok/s
Time to first answerBenchLeader43.3 s40.9 s
Context windowBenchLeader1,000,000 tokens1,048,576 tokens
Reasoning
GPQA DiamondEpoch AI Benchmarking Hub–88.1%†
Arena Hard PromptsArena (LMArena)–1453
LiveBench 2.0 ReasoningLiveBench82.9%†85.2%†
LiveBench 1.0 ReasoningLiveBench–83.9%†
Humanity's Last Exam (AA)Artificial Analysis48.0%35.0%
HLE-Diamond (AA)Artificial Analysis28.1%15.1%
CritPtArtificial Analysis22.3%10.6%
CritPt v2 hard subset (AA)Artificial Analysis32.5%24.5%
MysteryMechanismVals AI–12.2%†
Arena ExpertArena (LMArena)–1464
Chess PuzzlesEpoch AI Benchmarking Hub–33.0%†
Mystery Game PuzzlesEpoch AI Benchmarking Hub–17.0%†
LMCAEpoch AI Benchmarking Hub–35.8%
DTBenchEpoch AI Benchmarking Hub–88.2%
Coding
SciCodeSciCode–54.2%
FrontierCodeCognition–39.4%†
Arena CodingArena (LMArena)–1484
Arena WebDevArena (LMArena)15841565
LiveBench 2.0 CodingLiveBench68.8%†79.4%†
LiveBench 1.0 CodingLiveBench–77.2%†
SciCode (AA)Artificial Analysis53.4%54.2%
IOIVals AI–45.3%†
Code MigrationVals AI–30.6%†
SWE-bench MultimodalVals AI–33.1%†
Vibe Code Bench 1-100Vals AI–14.3%†
Show all 123 benchmarks
MetricPublisherMiniMax M3.1 FlashMistral Large 4
Coding
Vibe Code Bench v1.1Vals AI73.9%†78.4%†
Vibe Code Bench 1-100 v1.1Vals AI19.8%†–
SWE Atlas v1.1Scale AI SEAL–45.2%
SWE Atlas v1.1: Codebase QnAScale AI SEAL–61.3%
SWE Atlas v1.1: Test WritingScale AI SEAL–40.0%
SWE Atlas v1.1: RefactoringScale AI SEAL–34.3%
SWE-Bench Pro V2Scale AI SEAL96.6%†–
SWE-Bench Pro V2 HARDScale AI SEAL84.3%†–
MirrorCodeEpoch AI Benchmarking Hub–40.0%†
Surface Evolver BenchEpoch AI Benchmarking Hub–28.1%
ALE-BenchEpoch AI Benchmarking Hub–1213.2
WeirdML v3WeirdML–4.5%†
Terminal-Bench 4.0 (AA)Artificial Analysis25.8%26.8%
GameBench 2 by reasoning settingDuelLab GameBench 230.3†57.6†
Design Arena WebsiteDesign Arena12771258
FrontierCode 1.1 (AA)Artificial Analysis38.6%41.5%
FrontierCode 2.0 (AA)Artificial Analysis20.8%13.6%
Agents & tools
Terminal-BenchTerminal-Bench–27.6%†
APEX-AgentsMercor–51.6%†
Arena AgentArena (LMArena)-2.7-5.4
LiveBench Long AgenticLiveBench25.7%†24.7%†
LiveBench BrowserLiveBench80.6%†17.4%†
LiveBench Agentic CodingLiveBench–57.2%†
HiL-BenchScale AI SEAL65.7%–
GDPval-AA v2.1, normalised scoreArtificial Analysis47.9%46.1%
GDPval-AA v2.1Artificial Analysis14651431
Finance Agent v2Vals AI–54.7%†
Finance Agent v2.1Vals AI58.8%†61.4%†
Harvey's Legal Agent Benchmark v1.1Vals AI7.5%†17.5%†
Harvey's Legal Agent Benchmark v1Vals AI–15.8%†
Legal Research BenchVals AI34.1%†31.7%†
Tax Agent BenchVals AI32.1%†30.4%†
Terminal-Bench 4.0 (Vals)Vals AI25.3%†27.8%†
Arena Agent Task CompletionArena (LMArena)-0.7-4.7
AutomationBenchArtificial Analysis72.5%59.9%
GDP.pdf (AA)Artificial Analysis20.8%18.6%
Harvey LAB-AA v1.1Harvey–3.1%
AA-Briefcase v1.1Artificial Analysis14741393
GDPxGDPx by SemiAnalysis4.0%†–
Design Arena AgenticDesign Arena1252–
Terminal-Bench-Science 0.1 (AA)Artificial Analysis3.3%4.8%
CWE-Bench (AA)Artificial Analysis–50.8%
DeepSec-Bench (AA)Artificial Analysis–16.0%
CyberGym end-to-end (AA)Artificial Analysis–81.7%
MKA (AA)Artificial Analysis33.5%35.1%†
Maths
FrontierMath Tiers 1–3Epoch AI Benchmarking Hub–58.6%†
FrontierMath Tier 4Epoch AI Benchmarking Hub–26.8%†
OTIS Mock AIMEEpoch AI Benchmarking Hub–98.3%†
ProofBenchVals AI–10.0%
LiveBench MathematicsLiveBench–93.6%†
Arena MathsArena (LMArena)–1459
Knowledge
SimpleQA VerifiedEpoch AI Benchmarking Hub–20.0%†
LiveBench 2.0 Data AnalysisLiveBench71.4%†76.6%†
LiveBench ResearchLiveBench76.7%†80.3%†
LiveBench 1.0 Data AnalysisLiveBench–76.5%†
AA-Omniscience v1.1Artificial Analysis-6.4-7.5
PRBench FinanceScale AI SEAL45.6%–
PRBench LegalScale AI SEAL45.1%–
BioMysteryBenchVals AI–67.0%†
Excel Modeling BenchmarkVals AI–56.2%†
Excel Modeling Benchmark v1.1Vals AI67.8%†69.0%†
MedCodeVals AI–40.7%†
MedScribeVals AI–80.4%†
Access to Justice BenchmarkVals AI73.3%†–
Immigration BenchVals AI–63.7%†
AA-Omniscience v1.1: accuracyArtificial Analysis45.1%23.9%
AA-Omniscience v1.1: non-hallucinationArtificial Analysis6.3%58.7%
Instruction following
LiveBench LanguageLiveBench–49.6%†
LiveBench Instruction FollowingLiveBench–64.8%†
Arena Instruction FollowingArena (LMArena)–1418
Human preference
Arena TextArena (LMArena)–1420
Arena Creative WritingArena (LMArena)–1359
Arena Multi-turnArena (LMArena)–1429
Arena Longer QueriesArena (LMArena)–1420
Multimodal
Arena VisionArena (LMArena)–1244
LiveBench DesignLiveBench71.2%†76.3%†
MMMU-ProArtificial Analysis79.1%76.4%
SAGEVals AI85.3%†–
Furniture AssemblyEpoch AI Benchmarking Hub–20.8%†
VIALS-AAHandshake AI29.8%20.5%†
Code2Video-AAArtificial Analysis–931
AA-3DArtificial Analysis–882
Long context
AA-LCR v1.1Artificial Analysis79.7%81.3%
MLCRArtificial Analysis16.7%7.8%
Composite
Epoch Capabilities IndexEpoch AI Benchmarking Hub–154.6
LiveBench 2.0LiveBench63.7%†60.3%†
LiveBench 1.0LiveBench–71.8%†
AA Intelligence Index v5.0.1Artificial Analysis29.125.3
AA Intelligence Index v4.3.2Artificial Analysis42.738.4
AA Progress Index v1Artificial Analysis317.6312.4
Vals IndexVals AI53.8†–
Vals Finance IndexVals AI55.9%†56.9%†
Vals Router IndexVals AI53.8%–
Vals Legal IndexVals AI67.1%†65.6%†
AA Cyber IndexArtificial Analysis–49.5
Safety & honesty
FORTRESSScale AI SEAL52.1%23.1%
DistressBenchScale AI SEAL–77.0%
Metadata
AA Openness IndexArtificial Analysis27.838.9†

Data as of 2026-11-16. Best configuration of each model; every score links to its source on the model pages.

DuelLab GameBench 2, release gb2-20261111-7b598bb13de4 (https://benchmarks.duellab.org/releases/gb2-20261111-7b598bb13de4/), CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Scores mapped to BenchLeader model names; not endorsed by DuelLab.

What changed

  • MiniMax M3.1 Flash: BenchLeader Index 53.7 → 54.3.
  • Mistral Large 4: BenchLeader Index 52.2 → 52.9.
  • Mistral Large 4: 1 new result (GDPval-AA v2.1).
  • Mistral Large 4: 1 score revision.
  • Mistral Large 4: 2 score revisions.
  • MiniMax M3.1 Flash: BenchLeader Index 54.3 → 53.7.
  • MiniMax M3.1 Flash: 1 score revision.
  • Mistral Large 4: 6 score revisions.
  • Mistral Large 4: BenchLeader Index 52.7 → 52.2.
  • MiniMax M3.1 Flash: 6 score revisions.

Rankings: Best LLM overall · Best LLM for agents · Best cheap LLMs

MiniMax M3.1 Flash vs Mistral Large 4: questions

Is MiniMax M3.1 Flash better than Mistral Large 4?
MiniMax M3.1 Flash and Mistral Large 4 are neck and neck on quality (54.3 vs 52.9, within the ±6.4 margin). The BenchLeader Index combines every independent quality benchmark; the two are within the margin of error as of 16 Nov 2026, so check the category scores for your use.
MiniMax M3.1 Flash vs Mistral Large 4: which is cheaper, and will that change?
MiniMax M3.1 Flash is cheaper today: $0.262 against $1.05 per million tokens, blended at three input tokens per output token. No price change has been announced for either.
Is MiniMax M3.1 Flash or Mistral Large 4 better for coding?
MiniMax M3.1 Flash scores higher in coding (55 vs 54 on the category index, where 50 is average). On Arena WebDev, the top shared coding benchmark, it is 1584 for MiniMax M3.1 Flash against 1565 for Mistral Large 4.
Which is faster, MiniMax M3.1 Flash or Mistral Large 4?
MiniMax M3.1 Flash streams faster (151 vs 97 output tokens per second, Artificial Analysis 7-day median).
Is MiniMax M3.1 Flash or Mistral Large 4 better for agents and tool use?
MiniMax M3.1 Flash scores higher in agentic tasks (55 vs 49 on the category index, where 50 is average). On Arena Agent, the top shared agents & tools benchmark, it is -2.7 for MiniMax M3.1 Flash against -5.4 for Mistral Large 4.
More questions (4)
Which is better for long documents?
They are level on long-context benchmarks (65 vs 65). Both take about 1M tokens: MiniMax M3.1 Flash 1,000,000, Mistral Large 4 1,048,576.
Which has lower latency?
Mistral Large 4 answers first: 40.9 s against 43.3 s to the first answer token, as Artificial Analysis measures it. Reasoning effort drives most of this: a higher setting thinks longer before it answers.
Do independent benchmarks agree on MiniMax M3.1 Flash vs Mistral Large 4?
Across the 26 benchmarks both have, MiniMax M3.1 Flash leads 19, Mistral Large 4 leads 7. Every overall index that rates both puts MiniMax M3.1 Flash ahead.
When was MiniMax M3.1 Flash released?
MiniMax M3.1 Flash was released on 19 Oct 2026.