RankingMistral Large 4

Mistral Large 4

Compare models

Data updated 22 Nov 2026

Mistral Large 4 by Mistral is a Mistral Large model, released 2026-10-06. This profile collects the published benchmark results and reported effort settings for this model. Configurations are compared on the effort page. Ranks are on the leaderboard.

26 published benchmark measures · 64 measured effort results · 3 benchmark families contribute across 3 task areas. 2 capability estimates available in the full profile. See all results ↓ · Compare published benchmarks →

Model evidence summary

This profile combines published settings. It is not a runnable configuration or a leaderboard rank. Compare measured configurations →

Performance profile

Capabilities

Adjusted comparison score · 0–100
Mistral Large 4 capability radarComparison scores from 0 to 100. Unknown capabilities have no point; lines stop at gaps. Hollow points are preliminary. Exact values and support labels follow the chart.AgenticHard reasoningCodingHuman prefKnowledgeMultimodalLong context50100Mistral Large 4 · Agentic: 24.5 · PreliminaryMistral Large 4 · Coding: 69.8 · Preliminary

Filled points are supported; hollow points are preliminary. Lines stop at unknown capabilities. Exact values and sources follow below.

Agentic1 families · 1 with independent evidence · Preliminary24.5

1 core families; 42 direct opponents across 17 labs. Needs broader benchmark and opponent support

Estimated against the same complete panel: gpt-5.6-sol claude-opus-4-8 gemini-3.1-pro . Indirect comparisons assume performance can be summarized by one strength within this capability.

Refit without one family: No supported estimate; 13/13 scenarios unsupported. Without one publisher: No supported estimate; 15/15 unsupported. Smoothing check: No supported estimate; 3/3 unsupported. These are sensitivity checks, not confidence intervals.

Audit: family and publisher holdouts improved on neutral predictions within evaluated coverage; source sensitivity still requires inspection. Validation audit.

Matched family win shares below describe source evidence; they are not averaged to produce the adjusted score.

  • GDPval: 73.8 observed win share
    artificialanalysis.ai · Source 1
Hard reasoningNo comparable evidenceUnknown

0 core families; 0 direct opponents across 0 labs. No admitted core comparisons

Estimated against the same complete panel: gpt-5.6-sol claude-opus-4-8 gemini-3.1-pro . Indirect comparisons assume performance can be summarized by one strength within this capability.

Refit without one family: No supported estimate; 6/6 scenarios unsupported. Without one publisher: No supported estimate; 16/16 unsupported. Smoothing check: No supported estimate; 3/3 unsupported. These are sensitivity checks, not confidence intervals.

Audit: family and publisher holdouts improved on neutral predictions within evaluated coverage. Validation audit.

Matched family win shares below describe source evidence; they are not averaged to produce the adjusted score.

    Coding1 families · 1 with independent evidence · Preliminary69.8

    1 core families; 29 direct opponents across 9 labs. Needs broader benchmark and opponent support

    Estimated against the same complete panel: gpt-5.6-sol claude-opus-4-8 gemini-3.1-pro . Indirect comparisons assume performance can be summarized by one strength within this capability.

    Refit without one family: No supported estimate; 6/6 scenarios unsupported. Without one publisher: No supported estimate; 19/19 unsupported. Smoothing check: No supported estimate; 3/3 unsupported. These are sensitivity checks, not confidence intervals.

    Audit: family and publisher holdouts improved on neutral predictions within evaluated coverage. Validation audit.

    Matched family win shares below describe source evidence; they are not averaged to produce the adjusted score.

    • Terminal-Bench: 51.7 observed win share
      tbench.ai · Source 1
    Human pref1 families · 1 with independent evidence · PreliminaryUnknown

    1 core families; 75 direct opponents across 17 labs. No connected comparison to the complete reference panel

    Estimated against the same complete panel: gpt-5.6-sol claude-opus-4-8 gemini-3.1-pro . Indirect comparisons assume performance can be summarized by one strength within this capability.

    Refit without one family: No supported estimate; 1/1 scenarios unsupported. Without one publisher: No supported estimate; 1/1 unsupported. Smoothing check: No supported estimate; 3/3 unsupported. These are sensitivity checks, not confidence intervals.

    Audit: cross-family and cross-publisher predictions could not be evaluated from the available comparison network. Validation audit.

    Matched family win shares below describe source evidence; they are not averaged to produce the adjusted score.

    • Human preference (Arena): 52.0 observed win share
      lmarena.ai · Source 1
    KnowledgeNo comparable evidenceUnknown

    0 core families; 0 direct opponents across 0 labs. No admitted core comparisons

    Estimated against the same complete panel: gpt-5.6-sol claude-opus-4-8 gemini-3.1-pro . Indirect comparisons assume performance can be summarized by one strength within this capability.

    Refit without one family: No supported estimate; 4/4 scenarios unsupported. Without one publisher: No supported estimate; 5/5 unsupported. Smoothing check: No supported estimate; 3/3 unsupported. These are sensitivity checks, not confidence intervals.

    Audit: almost no publisher-held-out comparisons remained estimable. Cross-source validity is unresolved. Validation audit.

    Matched family win shares below describe source evidence; they are not averaged to produce the adjusted score.

      MultimodalNo comparable evidenceUnknown

      0 core families; 0 direct opponents across 0 labs. No admitted core comparisons

      Estimated against the same complete panel: gpt-5.6-sol claude-opus-4-8 gemini-3.1-pro . Indirect comparisons assume performance can be summarized by one strength within this capability.

      Refit without one family: No supported estimate; 8/8 scenarios unsupported. Without one publisher: No supported estimate; 8/8 unsupported. Smoothing check: No supported estimate; 3/3 unsupported. These are sensitivity checks, not confidence intervals.

      Audit: limited publisher-held-out coverage. Cross-source validity is unresolved. Validation audit.

      Matched family win shares below describe source evidence; they are not averaged to produce the adjusted score.

        Long contextNo comparable evidenceUnknown

        0 core families; 0 direct opponents across 0 labs. No admitted core comparisons

        Estimated against the same complete panel: gpt-5.6-sol claude-opus-4-8 gemini-3.1-pro . Indirect comparisons assume performance can be summarized by one strength within this capability.

        Refit without one family: No supported estimate; 5/5 scenarios unsupported. Without one publisher: No supported estimate; 5/5 unsupported. Smoothing check: No supported estimate; 3/3 unsupported. These are sensitivity checks, not confidence intervals.

        Audit: family and publisher holdouts underperformed neutral predictions. Treat this as an exploratory estimate. Validation audit.

        Matched family win shares below describe source evidence; they are not averaged to produce the adjusted score.

          Scores estimate outcomes against a shared reference panel; they are not accuracy percentages. Sparse or disconnected evidence cannot qualify an overall profile. Open a capability to inspect its evidence.

          Reported effort · Not specified

          Settings reported in this model's published benchmark results, including results outside the aggregate. Effort names are provider-specific. These are not API defaults or equal compute budgets.

          Mixed settings means multiple settings occur in the evidence. Best across efforts means the source selected its best reported result across settings; it does not mean Max. Unspecified settings stay unknown. This model-summary chart combines reported settings. The leaderboard keeps identified configurations separate and excludes unknown effort.

          Inspect each result and its source ↓ · Download effort evidence
          Score contributions and missing evidence

          3 contributing families across 3 capabilities. Fixed reference panels do not change when the catalog expands.

          Capability is fitted jointly across families. Capability estimates below describe different task areas; their weighted sum is not the Capability score.

          Results without reviewed compatibility or a reference match remain in the raw evidence below. Coverage counts only contributing results.

          Model information & shareable badge
          Lab
          Mistral
          Catalog status
          preview
          Availability
          Public preview API on Mistral Studio. Open weights promised by end of October 2026; weights and license are not yet established.
          Family
          Mistral Large
          Released
          2026-10-06
          Context
          1,000,000 tokens
          API list price
          $1.36 input / $4.18 output per million tokens
          Standard list prices per million tokens; cached input $0.14. Launch promotion is 50% off for two weeks, currently $0.68 input, $0.07 cached input and $2.09 output. Independent evaluators may use list prices. No named API effort scale established by the model card.
          License
          —
          Model card
          https://docs.mistral.ai/models/mistral-large-4-0
          Default Capability family coverage
          Documented-evidence family coverage/badge/mistral-large-4.svg

          Mistral Large 4 measured results by effort setting

          Each row is one published result for one effort setting. Evidence collected 2026-11-22. Compare the effort settings →

          BenchmarkEffortScoreSource
          AA-Briefcase EloHigh1347Artificial Analysis
          AA-Briefcase EloMedium1187Artificial Analysis
          AA-Briefcase EloReasoning1393Artificial Analysis
          AA-LCR v1.1Reasoning81.3%Artificial Analysis
          AA-Omniscience AccuracyHigh26.0%Artificial Analysis
          AA-Omniscience AccuracyMedium22.5%Artificial Analysis
          AA-Omniscience AccuracyReasoning23.9%Artificial Analysis
          AA-Omniscience IndexHigh5.1Artificial Analysis
          AA-Omniscience IndexMedium3.3Artificial Analysis
          AA-Omniscience IndexReasoning-7.5Artificial Analysis
          Artificial Analysis Cyber IndexUnspecified49.5Artificial Analysis
          AutomationBench-AAHigh57.8%Artificial Analysis
          AutomationBench-AAMedium52.8%Artificial Analysis
          AutomationBench-AAReasoning59.9%Artificial Analysis
          CritPtHigh10.0%Artificial Analysis
          CritPtMedium11.1%Artificial Analysis
          CritPtReasoning10.6%Artificial Analysis
          GDP.pdf: All-pass RateHigh17.6%Artificial Analysis
          GDP.pdf: All-pass RateMedium16.8%Artificial Analysis
          GDP.pdf: All-pass RateReasoning18.6%Artificial Analysis
          GDPval-AA v2.1 LeaderboardReasoning1431Artificial Analysis
          Harvey LAB-AA v1.1: Hallucination-Gated All-Pass RateReasoning3.1%Artificial Analysis
          Humanity's Last ExamReasoning35.0%Artificial Analysis
          LiveBench · AMPS Hard · 2026-06-25High98.0%LiveBench
          LiveBench · code completion · 2026-06-25High78.3%LiveBench
          LiveBench · code generation · 2026-06-25High76.1%LiveBench
          LiveBench · connections · 2026-06-25High48.0%LiveBench
          LiveBench · consecutive events · 2026-06-25High80.5%LiveBench
          LiveBench · integrals with game · 2026-06-25High90.0%LiveBench
          LiveBench · javascript · 2026-06-25High68.2%LiveBench
          LiveBench · logic with navigation · 2026-06-25High68.0%LiveBench
          LiveBench · math comp · 2026-06-25High97.1%LiveBench
          LiveBench · olympiad · 2026-06-25High89.3%LiveBench
          LiveBench · paraphrase · 2026-06-25High64.0%LiveBench
          LiveBench · plot unscrambling · 2026-06-25High40.9%LiveBench
          LiveBench · python · 2026-06-25High60.0%LiveBench
          LiveBench · simplify · 2026-06-25High66.2%LiveBench
          LiveBench · spatial · 2026-06-25High94.0%LiveBench
          LiveBench · story generation · 2026-06-25High67.5%LiveBench
          LiveBench · summarize · 2026-06-25High61.6%LiveBench
          LiveBench · tablejoin · 2026-06-25High49.1%LiveBench
          LiveBench · tablereformat · 2026-06-25High100.0%LiveBench
          LiveBench · theory of mind · 2026-06-25High80.8%LiveBench
          LiveBench · typescript · 2026-06-25High43.3%LiveBench
          LiveBench · typos · 2026-06-25High60.0%LiveBench
          LiveBench · zebra puzzle · 2026-06-25High93.0%LiveBench
          MLCR-AAHigh17.8%Artificial Analysis
          MLCR-AAReasoning7.8%Artificial Analysis
          MMMU-ProReasoning76.4%Artificial Analysis
          SciCodeReasoning54.2%Artificial Analysis
          sudo L7High11.7%Surge AI
          Terminal-Bench 4.0 · claudecodeMax27.9%Terminal-Bench
          Terminal-Bench v4.0High24.7%Artificial Analysis
          Terminal-Bench v4.0Medium23.2%Artificial Analysis
          Terminal-Bench v4.0Reasoning26.8%Artificial Analysis
          Terminal-Bench-Science 0.1High4.3%Artificial Analysis
          Terminal-Bench-Science 0.1Medium3.3%Artificial Analysis
          Terminal-Bench-Science 0.1Reasoning4.8%Artificial Analysis
          Vals AI · Code Migration v1High30.6%Vals AI
          Vals AI · Excel Modeling Benchmark v1High56.2%Vals AI
          Vals AI · Finance Agent v2High54.7%Vals AI
          Vals AI · Legal Research Bench v1High31.7%Vals AI
          Vals AI · Tax Agent Bench v1: Weighted PassHigh63.3%Vals AI
          Vals AI · Vibe Code Bench v1.1High78.4%Vals AI
          Benchmark scores & sources

          Original results, evaluation harnesses, and evidence behind this model.

          BenchmarkBucketScoreHarnessEvidenceSource-recorded date
          SWE-Atlas-QnASupporting evidence59.4%
          Reported settings & source

          Public Mistral Large4 Preview. Effort, sampling and evaluation budgets are not stated. Chart footnote attributes private Artificial Analysis evaluation before public harness launch; provider-republished claim, not a new independent result.

          Provider-published lab_self_report; retained as launch evidence, not an independent board admission. Narrative states59.4%; chart rounds to59. Exact printed chart label; asset https://mistral.ai/_astro/code-benchmarks---swe-atlas-qna-v3_2wiWq0.webp?dpl=6ac6a8353c2a68969e300fd2; asset SHA256 9f9d32e5c05b99d91047ead47225644319e95ed3ff0e51415c10681bffbb8d56.

          Comparison limit: Provider publication; exact benchmark-specific configuration and independent evaluation provenance require separate review before matched-board admission.

          Introducing Mistral Large 4 — public preview · Capabilities deep-dive; SWE-Atlas-QnA; chart 10 (mistral-chart-10.webp) · reviewed 2026-10-07

          Published configuration

          Effort: Not specified

          lab self-report

          Provider publication; exact benchmark-specific configuration and independent evaluation provenance require separate review before matched-board admission.

          Reviewed 2026-10-07
          Terminal-Bench · 4.0Coding28.3%
          Reported settings & source

          Public Mistral Large4 Preview. Effort, sampling and evaluation budgets are not stated. Chart footnote attributes private Artificial Analysis evaluation before public harness launch; provider-republished claim, not a new independent result.

          Provider-published lab_self_report; retained as launch evidence, not an independent board admission. Narrative states28.3%; chart rounds to28. Exact printed chart label; asset https://mistral.ai/_astro/code-benchmarks---terminal-bench-4%201-v3_Z2iGo8O.webp?dpl=6ac6a8353c2a68969e300fd2; asset SHA256 c3201abcd5c42890988e41c33719804fe09ff7ffe90bd0c71553e964a1f4d340.

          Comparison limit: Provider publication; exact benchmark-specific configuration and independent evaluation provenance require separate review before matched-board admission.

          Introducing Mistral Large 4 — public preview · Capabilities deep-dive; Terminal-Bench; chart 1 (mistral-chart-1.webp) · reviewed 2026-10-07

          Published configuration

          Effort: Not specified

          lab self-report

          Provider publication; exact benchmark-specific configuration and independent evaluation provenance require separate review before matched-board admission.

          Reviewed 2026-10-07
          Terminal-Bench 4.0Coding27.88%Terminal-Bench 4.0 reported

          Effort: Not specified

          contributes to capability
          official board2026-10-06
          ARC-AGI-2Hard reasoning————
          DeepSWE v1.1Agentic————
          GPQA DiamondHard reasoning————
          Humanity's Last ExamHard reasoning————
          LiveCodeBenchCoding————
          MMLU-ProKnowledge————
          OSWorld-VerifiedAgentic————
          SWE-bench ProAgentic————
          SWE-bench VerifiedAgentic————
          Terminal-Bench 2.1Agentic————