Noometry

Model comparison

Codestral vs Mistral Large 3

Mistral Large 3 is the stronger model overall, scoring 39.1 to 30.6 on the Noometry Index.

Last verified . 1 shared benchmarks.

Codestral Mistral AI

30.6

Rank #290 Reported

Mistral Large 3 Mistral AI

39.1

Rank #176 Confirmed

Summary

  • They share 1 benchmark with published results for both. Codestral scores higher in 1 category and Mistral Large 3 in 1 category; 2 gaps are clear of the uncertainty.
  • The widest gap is in coding, where Mistral Large 3 leads 34.4 to 27.3.
  • The biggest single-benchmark swing is Kagi LLM Benchmark: 32.5% for Codestral and 50.9% for Mistral Large 3.
  • Mistral Large 3 is cheaper at $0.25 / $0.75 per million input/output tokens, against $0.30 / $0.90 for Codestral.
  • Mistral Large 3 accepts more context: 262K tokens versus 256K.
  • Mistral Large 3 has downloadable open weights; the other is API-only.

Side by side

CodestralMistral Large 3
ProviderMistral AIMistral AI
Noometry Index30.639.1
Released2024-05-292025-12-02
WeightsProprietaryOpen
Context window256K262K
Max output8K8K
Input $ / M tokens$0.30$0.25
Output $ / M tokens$0.90$0.75
Results tracked724

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

Category by category

Coding Mistral Large 3 leads

Codestral: 27.3 (#321), Mistral Large 3: 34.4 (#237)

BenchmarkCodestralMistral Large 3
Aider Polyglot11.1%—
LMArena WebDev—1230
BigCodeBench Instruct41.8%—
LMArena Coding—1448
BigCodeBench Complete52.5%—
ALE-Bench137.78—
HumanEval+73.8%—
MBPP+61.9%—

Reasoning Codestral leads

Codestral: 19.8 (#251), Mistral Large 3: 15.2 (#319)

BenchmarkCodestralMistral Large 3
Kagi LLM Benchmark32.5%50.9%
NYT Connections (extended)—7.5%
Thematic Generalization—23%
LMArena Hard Prompts—1429

Math Not comparable

Codestral: —, Mistral Large 3: 38.7 (#129)

BenchmarkCodestralMistral Large 3
LMArena Math—1414

Knowledge Not comparable

Codestral: —, Mistral Large 3: 36.0 (#177)

BenchmarkCodestralMistral Large 3
Vectara Hallucination Rate—14.5%
LMArena Expert—1421

Multimodal Not comparable

Codestral: —, Mistral Large 3: 38.2 (#66)

BenchmarkCodestralMistral Large 3
LMArena Vision—1221

Multilingual Not comparable

Codestral: —, Mistral Large 3: 52.5 (#84)

BenchmarkCodestralMistral Large 3
LMArena Non-English—1413
LMArena Chinese—1447
LMArena French—1455
LMArena German—1437
LMArena Japanese—1394
LMArena Korean—1384
LMArena Russian—1411
LMArena Spanish—1440

Instruction Following Not comparable

Codestral: —, Mistral Large 3: 74.0 (#108)

BenchmarkCodestralMistral Large 3
LMArena Instruction Following—1403

Long Context Not comparable

Codestral: —, Mistral Large 3: 43.1 (#105)

BenchmarkCodestralMistral Large 3
LMArena Longer Query—1413

Writing & Preference Not comparable

Codestral: —, Mistral Large 3: 60.0 (#101)

BenchmarkCodestralMistral Large 3
LMArena Text—1428
LMArena Creative Writing—1386
EQ-Bench Creative Writing—1412
LMArena Multi-Turn—1429

Frequently asked questions

Is Codestral better than Mistral Large 3?

Mistral Large 3 is the stronger model overall, scoring 39.1 to 30.6 on the Noometry Index.

Which is cheaper, Codestral or Mistral Large 3?

Mistral Large 3 is cheaper. It lists at $0.25 per million input tokens and $0.75 per million output tokens; Codestral lists at $0.30 and $0.90.

Is Codestral or Mistral Large 3 better for coding?

Mistral Large 3 scores higher on coding benchmarks: 34.4 versus 27.3 in the Noometry coding category.

Which has the bigger context window?

Mistral Large 3 does, with 262K tokens against 256K.

How many benchmarks do Codestral and Mistral Large 3 share?

1 benchmark has published results for both models. Codestral has 7 scored results on Noometry and Mistral Large 3 has 24.

Related comparisons

Go deeper