Model comparison

Codestral vs DeepSeek-V3.1

DeepSeek-V3.1 is the stronger model overall, scoring 42.8 to 30.6 on the Noometry Index.

Last verified October 11, 2026. 1 shared benchmarks.

Codestral

Mistral AI

30.6

Rank #290 Reported

DeepSeek-V3.1

DeepSeek

42.8

Rank #108 Confirmed

Summary

They share 1 benchmark with published results for both. Codestral scores higher in 0 categories and DeepSeek-V3.1 in 2 categories; 2 gaps are clear of the uncertainty.
The widest gap is in coding, where DeepSeek-V3.1 leads 40.3 to 27.3.
The biggest single-benchmark swing is Kagi LLM Benchmark: 32.5% for Codestral and 53.2% for DeepSeek-V3.1.
DeepSeek-V3.1 is cheaper at $0.25 / $0.95 per million input/output tokens, against $0.30 / $0.90 for Codestral.
Codestral accepts more context: 256K tokens versus 164K.
DeepSeek-V3.1 has downloadable open weights; the other is API-only.

Side by side

Codestral and DeepSeek-V3.1 specifications
	Codestral	DeepSeek-V3.1
Provider	Mistral AI	DeepSeek
Noometry Index	30.6	42.8
Released	2024-05-29	2025-08-21
Weights	Proprietary	Open
Context window	256K	164K
Max output	8K	8K
Input $ / M tokens	$0.30	$0.25
Output $ / M tokens	$0.90	$0.95
Results tracked	7	27

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

Category by category

Coding DeepSeek-V3.1 leads

Codestral: 27.3 (#321), DeepSeek-V3.1: 40.3 (#144)

Coding benchmarks
Benchmark	Codestral	DeepSeek-V3.1
Aider Polyglot	11.1%	—
WeirdML	—	38.4%
BigCodeBench Instruct	41.8%	—
LMArena Coding	—	1417
BigCodeBench Complete	52.5%	—
ALE-Bench	137.78	—
HumanEval+	73.8%	—
MBPP+	61.9%	—

Reasoning DeepSeek-V3.1 leads

Codestral: 19.8 (#251), DeepSeek-V3.1: 27.9 (#110)

Reasoning benchmarks
Benchmark	Codestral	DeepSeek-V3.1
Kagi LLM Benchmark	32.5%	53.2%
SimpleBench	—	40%
LMArena Hard Prompts	—	1417
DTBench	—	82.7%
LMCA	—	24.3%
Epoch Capabilities Index	—	139.92
ForecastBench	—	58

Math Not comparable

Codestral: —, DeepSeek-V3.1: 38.9 (#122)

Math benchmarks
Benchmark	Codestral	DeepSeek-V3.1
LMArena Math	—	1420

Knowledge Not comparable

Codestral: —, DeepSeek-V3.1: 43.7 (#90)

Knowledge benchmarks
Benchmark	Codestral	DeepSeek-V3.1
Vectara Hallucination Rate	—	5.5%
LMArena Expert	—	1405

Multilingual Not comparable

Codestral: —, DeepSeek-V3.1: 51.6 (#106)

Multilingual benchmarks
Benchmark	Codestral	DeepSeek-V3.1
LMArena Non-English	—	1400
LMArena Chinese	—	1469
LMArena French	—	1447
LMArena German	—	1411
LMArena Japanese	—	1378
LMArena Korean	—	1337
LMArena Russian	—	1405
LMArena Spanish	—	1431

Instruction Following Not comparable

Codestral: —, DeepSeek-V3.1: 73.9 (#110)

Instruction Following benchmarks
Benchmark	Codestral	DeepSeek-V3.1
LMArena Instruction Following	—	1400

Long Context Not comparable

Codestral: —, DeepSeek-V3.1: 36.3 (#232)

Long Context benchmarks
Benchmark	Codestral	DeepSeek-V3.1
Fiction.LiveBench	—	52.8%
LMArena Longer Query	—	1422

Writing & Preference Not comparable

Codestral: —, DeepSeek-V3.1: 60.3 (#98)

Writing & Preference benchmarks
Benchmark	Codestral	DeepSeek-V3.1
LMArena Text	—	1420
LMArena Creative Writing	—	1401
EQ-Bench Creative Writing	—	1436
LMArena Multi-Turn	—	1408

Frequently asked questions

Is Codestral better than DeepSeek-V3.1?

DeepSeek-V3.1 is the stronger model overall, scoring 42.8 to 30.6 on the Noometry Index.

Which is cheaper, Codestral or DeepSeek-V3.1?

DeepSeek-V3.1 is cheaper. It lists at $0.25 per million input tokens and $0.95 per million output tokens; Codestral lists at $0.30 and $0.90.

Is Codestral or DeepSeek-V3.1 better for coding?

DeepSeek-V3.1 scores higher on coding benchmarks: 40.3 versus 27.3 in the Noometry coding category.

Which has the bigger context window?

Codestral does, with 256K tokens against 164K.

How many benchmarks do Codestral and DeepSeek-V3.1 share?

1 benchmark has published results for both models. Codestral has 7 scored results on Noometry and DeepSeek-V3.1 has 27.