Noometry

Mistral AI, open weights

Mistral Large 3

Mistral Large 3 by Mistral AI ranks 176th of 354 ranked models on the Noometry Index as of October 2026, with a score of 39.1. Its strongest category is multimodal, where it ranks 66th. API pricing starts at $0.25 per million input tokens and $0.75 per million output tokens, with a 262K-token context window.

Last verified

Noometry rank
#176 of 354
Index score
39.1
Evidence
Confirmed 24 results
Provider
Mistral AI
Released
December 2, 2025
Weights
Open weights
Reasoning
No
Context window
262K
Max output
8K
Input price
$0.25 / M
Output price
$0.75 / M
Blended price
$0.38 / M
Output speed
7 tokens/s Kagi
Value
#57 of 219
Knowledge cutoff
November 2024
Input
text, image

Category scores

Each category score combines every public result we have in that category.

  1. Coding 34.4
  2. Reasoning 15.2
  3. Math 38.7
  4. Knowledge 36.0
  5. Multimodal 38.2
  6. Multilingual 52.5
  7. Instruction Following 74.0
  8. Long Context 43.1
  9. Writing & Preference 60.0
CategoryScoreRankResults
Coding34.4#2372
Reasoning15.2#3194
Math38.7#1291
Knowledge36.0#1772
Multimodal38.2#661
Multilingual52.5#841
Instruction Following74.0#1081
Long Context43.1#1051
Writing & Preference60.0#1014

Strengths and weaknesses

Categories where Mistral Large 3 places highest and lowest among the models ranked in each, with its score against that category's median.

Strongest categories

CategoryScorevs medianRank
Multilingual52.5+5.1#84 of 297, top 29%
Writing & Preference60.0+6.2#101 of 312, top 33%
Instruction Following74.0+2.8#108 of 305, top 36%

Weakest categories

CategoryScorevs medianRank
Reasoning15.2−8.4#319 of 350, top 92%
Coding34.4−4.3#237 of 340, top 70%
Knowledge36.0−1.3#177 of 314, top 57%

Closest competitors

The models ranked just above and below Mistral Large 3. When scores are this close, price and speed are often the better way to choose.

ModelRankScoreBlended $/MSpeed
Qwen3 32B#17239.2$1.2286Compare
Gemini 2.0 Pro#17339.1——Compare
Molmo 2 8b#17439.1——Compare
Mercury 2#17539.1$0.38—Compare
GLM-4.5-Air#17738.9$0.43160Compare
MiniMax-M2.1#17838.9$0.52—Compare
Qwen3-30B-A3B#17938.9$0.2142Compare
GLM-4.7-Flash#18038.8$0.15—Compare

Sponsored placements are available on pages like this one. Advertise on Noometry

Benchmark results

Every published result we track, with its source. Bold rows are the ones used for ranking; where several exist we prefer independent runs over self-reported numbers.

Coding

BenchmarkScorePositionSettingSourceDate
LMArena WebDev1230#106 of 113, top 94%LMArena2026-10-08
LMArena Coding1448#88 of 294, top 30%LMArena2026-10-08

Reasoning

BenchmarkScorePositionSettingSourceDate
Kagi LLM Benchmark50.9%#63 of 99, top 64%Kagi LLM Benchmark
NYT Connections (extended)7.5%#90 of 91, top 99%Lech Mazur benchmarks
Thematic Generalization23%#22 of 23, top 96%Lech Mazur benchmarks
LMArena Hard Prompts1429#89 of 297, top 30%LMArena2026-10-08

Math

BenchmarkScorePositionSettingSourceDate
LMArena Math1414#110 of 285, top 39%LMArena2026-10-08

Knowledge

BenchmarkScorePositionSettingSourceDate
Vectara Hallucination Rate (lower is better)14.5%#85 of 96, top 89%Vectara Hallucination Leaderboard
LMArena Expert1421#107 of 273, top 40%LMArena2026-10-08

Multimodal

BenchmarkScorePositionSettingSourceDate
LMArena Vision1221#71 of 122, top 59%LMArena2026-10-09

Multilingual

BenchmarkScorePositionSettingSourceDate
LMArena Non-English1413#84 of 297, top 29%LMArena2026-10-08
LMArena Chinese1447#105 of 285, top 37%LMArena2026-10-08
LMArena French1455#63 of 223, top 29%LMArena2026-10-08
LMArena German1437#61 of 231, top 27%LMArena2026-10-08
LMArena Japanese1394#64 of 211, top 31%LMArena2026-10-08
LMArena Korean1384#72 of 213, top 34%LMArena2026-10-08
LMArena Russian1411#92 of 283, top 33%LMArena2026-10-08
LMArena Spanish1440#67 of 226, top 30%LMArena2026-10-08

Instruction Following

BenchmarkScorePositionSettingSourceDate
LMArena Instruction Following1403#102 of 298, top 35%LMArena2026-10-08

Long Context

BenchmarkScorePositionSettingSourceDate
LMArena Longer Query1413#100 of 291, top 35%LMArena2026-10-08

Writing & Preference

BenchmarkScorePositionSettingSourceDate
LMArena Text1428#83 of 297, top 28%LMArena2026-10-08
LMArena Creative Writing1386#100 of 295, top 34%LMArena2026-10-08
EQ-Bench Creative Writing1412#69 of 115, top 60%EQ-Bench
LMArena Multi-Turn1429#83 of 295, top 29%LMArena2026-10-08

API pricing by provider

RouteInput $/MOutput $/MCached input $/MChecked
bedrock$0.50$1.50—2026-10-10
openrouter$0.25$0.75$0.0252026-10-10

All Mistral AI API prices → Estimate your cost →

Compare Mistral Large 3

Other Mistral AI models

Frequently asked questions

How good is Mistral Large 3?

Mistral Large 3 by Mistral AI ranks 176th of 354 ranked models on the Noometry Index as of October 2026, with a score of 39.1. Its strongest category is multimodal, where it ranks 66th. API pricing starts at $0.25 per million input tokens and $0.75 per million output tokens, with a 262K-token context window.

How much does Mistral Large 3 cost?

Mistral Large 3 costs $0.25 per million input tokens and $0.75 per million output tokens on openrouter, with cached input at $0.025.

What is Mistral Large 3's context window?

Mistral Large 3 accepts up to 262K tokens of input and can write up to 8K tokens in one response.

Is Mistral Large 3 open source?

Yes. Mistral Large 3's weights are downloadable; check the license for commercial terms.

How fast is Mistral Large 3?

Mistral Large 3 generated about 7 output tokens per second in the Kagi LLM Benchmark's timed runs. Speed varies by provider, load and reasoning effort.

What are Mistral Large 3's strengths and weaknesses?

Relative to other ranked models, Mistral Large 3 places best in multilingual, writing & preference, instruction following and lowest in reasoning, coding, knowledge.

What is Mistral Large 3 best at?

Its best category is multimodal, where it ranks 66th on Noometry.