Skip to content
Model Pareto

Mistral Large 4 vs Mistral Large 3

Mistral Large 4 scores higher than Mistral Large 3 in both shared categories. Mistral Large 4 costs 2.8× as much as Mistral Large 3 ($2.06 vs $0.75 per million tokens, blended 3:1 input:output).
Mistral Large 4Mistral Large 3
Focus

Coding: quality vs price

Compared models are ringed; the other 117 ranked here are greyed.

Lower price is better. Pareto frontier: Nex-N2.5-Mini, Nex-N2.5-Pro, MiMo-V2.6-Flash, Ternary Bonsai 2 27B, MiMo-V2.6-Pro, Gemini 4 Argon, Claude Opus 5.5. 34 models are not plotted: GLM-5.3 Flash NVFP4, DeepSeek V4.1 Flash NVFP4, GLM-5.3 NVFP4, DeepSeek-V4-Pro-0813-nvfp4-DSpark, Xing4.0-29B-A4B, Ternary Bonsai 27B, Atria Dawn Preview, Maple-Preview, Bonsai 27B, Qwen3.8 27B NVFP4, Intern-S2-397B, Phi-4-reasoning-plus, Solar Open2 250B, Laguna S 2.1, Motif 3, Nex-N2-Pro, K-EXAONE 2.0 0803, K2 Horizon 375B A23B, A.X-K2, K2 Horizon MoVA 36B A4B, Intern-S2-Preview (35B-A3B), North Mini Code, G9v3-39A5B, Falcon-H1-34B-Instruct, Qwen3.8 Flash-Next NVFP4, Devstral 2, K2 Horizon 7B, K2-Horizon-32B, Devstral Small 2, MiniCPM5-2B, Ling 3.0 Tiny, K2 Horizon 3.7B, K2 Horizon 0.9B, LFM2.5-2.6B.

34 models with no price data — shown in the strip at the left edge

GLM-5.3 Flash NVFP4, DeepSeek V4.1 Flash NVFP4, GLM-5.3 NVFP4, DeepSeek-V4-Pro-0813-nvfp4-DSpark, Xing4.0-29B-A4B, Ternary Bonsai 27B, Atria Dawn Preview, Maple-Preview, Bonsai 27B, Qwen3.8 27B NVFP4, Intern-S2-397B, Phi-4-reasoning-plus, Solar Open2 250B, Laguna S 2.1, Motif 3, Nex-N2-Pro, K-EXAONE 2.0 0803, K2 Horizon 375B A23B, A.X-K2, K2 Horizon MoVA 36B A4B, Intern-S2-Preview (35B-A3B), North Mini Code, G9v3-39A5B, Falcon-H1-34B-Instruct, Qwen3.8 Flash-Next NVFP4, Devstral 2, K2 Horizon 7B, K2-Horizon-32B, Devstral Small 2, MiniCPM5-2B, Ling 3.0 Tiny, K2 Horizon 3.7B, K2 Horizon 0.9B, LFM2.5-2.6B

  • Best-value frontier (nothing is both cheaper and better)
  • Evidencestrong → weak
  • Estimated from other categories
  • Mistral Large 4
    Quality
    62.5
    Rank
    #25/119
    Price
    $2.06/M tok
    Speed
    —
    In$1.36Out$4.18/M tok

    vs Mistral Large 3: +38.4 quality · 2.8× price

  • Mistral Large 3
    Quality
    24.1
    Rank
    #98/119
    Price
    $0.75/M tok
    Speed
    77tok/s
    In$0.50▾Out$1.50▾/M tok

    vs Mistral Large 4: −38.4 quality · 0.36× price

Overview

Metric
Mistral Large 4
Mistral Large 3
Lab
Mistral AI
Mistral AI
Released
Oct 6, 2026
Dec 2, 2025
Weights
Proprietary
Open
Input price
$/M tok
$1.36
$0.50Best
Output price
$/M tok
$4.18
$1.50Best
Blended price
$/M tok · 3:1 in:out
$2.065
$0.75Best
Output speed
tok/s
—
77
Context
tokens
—
256K

Quality by category

0–100 within each category (100 = best tracked model). Rank is among all models ranked there. Click a row to focus it.

Metric
Mistral Large 4
Mistral Large 3
Coding
62.5±6.5Best
#25/119Lab-reported
24.1±6.4
#98/119Verified
Writing & chat
Not ranked
38.7±3.3
#83/156Verified
Vision
Not ranked
27.6±4.8
#73/91Verified
Hard reasoning
Not ranked
25.0±3.4
#140/200Verified
Agents
81.2±10.1Best
#14/178Lab-reported
16.4±4.2
#157/178Verified

Coding benchmarks

Rank among models with a published score, and the leaderboard around each model. ~ italic = no published score, estimated from related benchmarks.

Metric
Mistral Large 4
Mistral Large 3
SWE-bench Verified
day 0
~#2/14~87.0%Estimated
  1. #1Claude Opus 596.0%
  2. ~#2Mistral Large 4Estimated87.0%
  3. #2Claude Sonnet 585.2%
estimated from related benchmarks
~#12/14~69.2%Estimated
  1. #1Claude Opus 596.0%
  2. #11Nemotron 3 Ultra70.7%
  3. ~#12Mistral Large 3Estimated69.2%
  4. #12Intern-S2-Preview (35B-A3B)64.0%
estimated from related benchmarks
SWE-bench Pro (Public)
day 0
~#6/20~68.0%Estimated
  1. #1Claude Opus 5.589.9%
  2. #5Intern-S2-397B68.5%
  3. ~#6Mistral Large 4Estimated68.0%
  4. #6Hy4 preview65.7%
estimated from related benchmarks
~#19/20~47.2%Estimated
  1. #1Claude Opus 5.589.9%
  2. #18Muse Glimmer51.2%
  3. ~#19Mistral Large 3Estimated47.2%
  4. #19Nex-N2.5-Mini43.8%
estimated from related benchmarks
DeepSWE v1.1
day 0
#23/3461.7%Lab-reported
  1. #1Gemini 4 Argon78.8%
  2. #22Claude Opus 562.5%
  3. #23Mistral Large 461.7%
  4. #24GLM-5.361.4%
mistral.ai · 2026-10-07
~#31/35~43.2%Estimated
  1. #1Gemini 4 Argon78.8%
  2. #30Qwen3.8 Max (0902)51.0%
  3. ~#31Mistral Large 3Estimated43.2%
  4. #31Qwen3.8 27B42.2%
estimated from related benchmarks
Terminal-Bench 4.0
day 0
#17/8828.3%BestLab-reported
  1. #1Claude Sonnet 5.563.6%
  2. #15GLM-5.3 Flash32.8%
  3. #17Mistral Large 428.3%
  4. #18DeepSeek V4.1 Flash26.8%
mistral.ai · 2026-10-07
#53/880.0%Verified
  1. #1Claude Sonnet 5.563.6%
  2. #53Ministral 3 8B0.0%
  3. #53Mistral Large 30.0%
  4. #53Mistral Medium 3.50.0%
artificialanalysis.ai · 2026-09-24
LiveCodeBench
third-party
~#6/25~88.5%Estimated
  1. #1Qwen3.8-Omni-Flash92.6%
  2. #5Nemotron 3 Ultra89.0%
  3. ~#6Mistral Large 4Estimated88.5%
  4. #6gpt-oss-120b87.8%
estimated from related benchmarks
#17/2446.5%Verified
  1. #1Qwen3.8-Omni-Flash92.6%
  2. #16Falcon-H1-34B-Instruct49.7%
  3. #17Mistral Large 346.5%
  4. #18Devstral 244.8%
artificialanalysis.ai · 2026-09-24
SciCode
third-party
~#27/106~53.3%Estimated
  1. #1Claude Opus 5.566.9%
  2. #26DeepSeek-V4-Pro-0813-nvfp4-DSpark53.8%
  3. ~#27Mistral Large 4Estimated53.3%
  4. #27Grok 4.653.0%
estimated from related benchmarks
#80/10536.6%Verified
  1. #1Claude Opus 5.566.9%
  2. #79G9v3-39A5B36.8%
  3. #80Mistral Large 336.6%
  4. #80Qwen3.6 35B A3B36.6%
artificialanalysis.ai · 2026-09-24
AA Coding Agent Index
third-party
#12/1749.8Lab-reported
  1. #1Claude Opus 5.566.0
  2. #11Kimi K351.9
  3. #12Mistral Large 449.8
  4. #13Grok 4.647.0
mistral.ai · 2026-10-07
~last/18below measured range (<41.1)Estimated
  1. #1Claude Opus 5.566.0
  2. #17GPT-6 Luna41.1
  3. ~#18Mistral Large 3Estimated<41.1
estimated from related benchmarks
Benchmarks in other categories (4)

Writing & chat benchmarks

Rank among models with a published score, and the leaderboard around each model. ~ italic = no published score, estimated from related benchmarks.

Metric
Mistral Large 4
Mistral Large 3
IFBench
day 0
—
#91/13336.2%Verified
  1. #1MiniMax-M382.9%
  2. #90Command A36.5%
  3. #91Mistral Large 336.2%
  4. #92Nanbeige4.1-3B35.4%
artificialanalysis.ai · 2026-09-24
MMMLU
day 0
—
~#4/5~81.5%Estimated
  1. #1Gemini 3.1 Pro92.6%
  2. #3Claude Haiku 4.583.0%
  3. ~#4Mistral Large 3Estimated81.5%
  4. #4gpt-oss-120b81.3%
estimated from related benchmarks
LMArena Text, Non-English (Elo)
third-party
—
#39/561397Verified
  1. #1Gemini 4 ArgonFew votes1512
  2. #38Nemotron 3 Ultra1405
  3. #39Mistral Large 31397
  4. #39Qwen3.5 122B A10B1397
lmarena.ai · 2026-09-24
LMArena Text (Elo)
third-party
—
#41/561413Verified
  1. #1Gemini 4 Argon1525
  2. #40Claude Haiku 4.51415
  3. #41Mistral Large 31413
  4. #42GPT-5.4 nano1402
lmarena.ai · 2026-09-24
EQ-Bench Creative Writing v3 (Elo)
third-party
—
#22/251412Verified
  1. #1GPT-6 Astra2173
  2. #21Gemini 3.1 Pro1491
  3. #22Mistral Large 31412
  4. #23Gemma 4 31B1368
eqbench.com · 2026-09-24

Vision benchmarks

Rank among models with a published score, and the leaderboard around each model. ~ italic = no published score, estimated from related benchmarks.

Metric
Mistral Large 4
Mistral Large 3
MMMU
day 0
—
~#2/3~71.3%Estimated
  1. #1Claude Haiku 4.573.2%
  2. ~#2Mistral Large 3Estimated71.3%
  3. #2Llama 4 Maverick69.4%
estimated from related benchmarks
MMMU-Pro
day 0
—
#64/8555.7%Verified
  1. #1Claude Opus 5.587.7%
  2. #63Mistral Small 456.8%
  3. #64Mistral Large 355.7%
  4. #65Magistral Small 1.255.5%
artificialanalysis.ai · 2026-09-24
CharXiv Reasoning
day 0
—
~last/9below measured range (<74.5%)Estimated
  1. #1Gemini 3.8 Flash86.2%
  2. #8Gemini 3.5 Flash-Lite74.5%
  3. ~#9Mistral Large 3Estimated<74.5%
estimated from related benchmarks
OmniDocBench v1.5
third-party
—
~last/8below measured range (<75.8%)Estimated
  1. #1MiniMax-M391.6%
  2. #7Muse Glimmer75.8%
  3. ~#8Mistral Large 3Estimated<75.8%
estimated from related benchmarks
GDP.pdf (AA)
third-party
—
#37/431.2%Verified
  1. #1GPT-6.1 Sol32.0%
  2. #37Ministral 3 14B1.2%
  3. #37Mistral Large 31.2%
  4. #39Command A+1.0%
artificialanalysis.ai · 2026-09-24
Chartography
third-party
—
#23/239.0%Verified
  1. #1GPT-6 Astra71.0%
  2. #22Inkling10.0%
  3. #23Mistral Large 39.0%
surgehq.ai · 2026-09-24

Hard reasoning benchmarks

Rank among models with a published score, and the leaderboard around each model. ~ italic = no published score, estimated from related benchmarks.

Metric
Mistral Large 4
Mistral Large 3

Agents benchmarks

Rank among models with a published score, and the leaderboard around each model. ~ italic = no published score, estimated from related benchmarks.

Metric
Mistral Large 4
Mistral Large 3
τ²-bench (Telecom)
day 0
~#12/118~91.7%Estimated
  1. #1Step 3.7 Flash98.5%
  2. #11Ring-2.6-1T92.4%
  3. ~#12Mistral Large 4Estimated91.7%
  4. #12MiMo-V2.590.6%
estimated from related benchmarks
#71/11724.6%Verified
  1. #1Step 3.7 Flash98.5%
  2. #69Ministral 3 3B24.9%
  3. #71Mistral Large 324.6%
  4. #72Devstral Small 223.4%
artificialanalysis.ai · 2026-09-24
OSWorld-Verified
day 0
~#3/12~83.4%Estimated
  1. #1Kimi K384.8%
  2. #2Qwen3.8 27B84.3%
  3. ~#3Mistral Large 4Estimated83.4%
  4. #3Nex-N2.5-Pro82.2%
estimated from related benchmarks
~#11/12~60.6%Estimated
  1. #1Kimi K384.8%
  2. #10Muse Glimmer65.9%
  3. ~#11Mistral Large 3Estimated60.6%
  4. #11Claude Haiku 4.550.7%
estimated from related benchmarks
OSWorld 2.0
day 0
~#3/5~55.6%Estimated
  1. #1GLM-5.3 Flash59.1%
  2. #2Kimi K358.3%
  3. ~#3Mistral Large 4Estimated55.6%
  4. #3GPT-5.6 Terra50.2%
estimated from related benchmarks
~last/5below measured range (<47.5%)Estimated
  1. #1GLM-5.3 Flash59.1%
  2. #4GPT-5.547.5%
  3. ~#5Mistral Large 3Estimated<47.5%
estimated from related benchmarks
BrowseComp
day 0
~#5/18~90.5%Estimated
  1. #1Atria Dawn Preview92.5%
  2. #4Claude Opus 590.8%
  3. ~#5Mistral Large 4Estimated90.5%
  4. #5Nex-N2.5-Pro89.7%
estimated from related benchmarks
~#15/18~49.1%Estimated
  1. #1Atria Dawn Preview92.5%
  2. #14Inkling77.1%
  3. ~#15Mistral Large 3Estimated49.1%
  4. #15Mistral Medium 3.548.6%
estimated from related benchmarks
GDPval-AA (Elo)
third-party
~#14/88~1585Estimated
  1. #1Claude Opus 5.51866
  2. #13Qwen3.8 2.4T A95B1597
  3. ~#14Mistral Large 4Estimated1585
  4. #14GPT-6.1 Sol1575
estimated from related benchmarks
#74/87406Verified
  1. #1Claude Opus 5.51866
  2. #73Qwen3.5 9B407
  3. #74Mistral Large 3406
  4. #75Nova 2.0 Lite354
artificialanalysis.ai · 2026-09-24
AutomationBench-AA
third-party
#10/8159.9%BestLab-reported
  1. #1Claude Opus 5.569.5%
  2. #10Gemini 3.8 Flash59.9%
  3. #10Mistral Large 459.9%
  4. #12GPT-5.6 Terra59.6%
mistral.ai · 2026-10-07
#62/811.5%Verified
  1. #1Claude Opus 5.569.5%
  2. #62Granite 4.2 30B1.5%
  3. #62Mistral Large 31.5%
  4. #62North Mini Code1.5%
artificialanalysis.ai · 2026-09-24
τ-Bench Banking (AA)
third-party
~#12/90~43.2%Estimated
  1. #1Muse Spark 1.350.5%
  2. #11Grok 4.643.3%
  3. ~#12Mistral Large 4Estimated43.2%
  4. #12Claude Opus 542.1%
estimated from related benchmarks
#75/895.8%Verified
  1. #1Muse Spark 1.350.5%
  2. #73NVIDIA Nemotron 3 Nano 30B A3B6.0%
  3. #75Mistral Large 35.8%
  4. #75Trinity Large Thinking5.8%
artificialanalysis.ai · 2026-09-24