Model comparison

Codestral vs GLM-4.5

GLM-4.5 is the stronger model overall, scoring 42.0 to 30.6 on the Noometry Index. Codestral costs 2.2× less per token, which makes it the better buy when GLM-4.5's lead doesn't matter for your workload.

Last verified October 10, 2026. 2 shared benchmarks.

Codestral

Mistral AI

30.6

Rank #290 Reported

GLM-4.5 Z.ai (Zhipu)

42.0

Rank #122 Confirmed

Summary

They share 2 benchmarks with published results for both. Codestral scores higher in 0 categories and GLM-4.5 in 2 categories; 2 gaps are clear of the uncertainty.
The widest gap is in coding, where GLM-4.5 leads 41.4 to 27.3.
The biggest single-benchmark swing is Kagi LLM Benchmark: 32.5% for Codestral and 57.9% for GLM-4.5.
Codestral is cheaper at $0.30 / $0.90 per million input/output tokens, against $0.60 / $2.20 for GLM-4.5.
Codestral accepts more context: 256K tokens versus 131K.
GLM-4.5 has downloadable open weights; the other is API-only.

Side by side

Codestral and GLM-4.5 specifications
	Codestral	GLM-4.5
Provider	Mistral AI	Z.ai (Zhipu)
Noometry Index	30.6	42.0
Released	2024-05-29	2025-07-27
Weights	Proprietary	Open
Context window	256K	131K
Max output	8K	98K
Input $ / M tokens	$0.30	$0.60
Output $ / M tokens	$0.90	$2.20
Results tracked	7	27

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

Category by category

Coding GLM-4.5 leads

Codestral: 27.3 (#321), GLM-4.5: 41.4 (#125)

Coding benchmarks
Benchmark	Codestral	GLM-4.5
ALE-Bench	137.78	344.82
SWE-bench Verified (bash only)	—	54.2%
Aider Polyglot	11.1%	—
WeirdML	—	40.6%
BigCodeBench Instruct	41.8%	—
LMArena Coding	—	1434
BigCodeBench Complete	52.5%	—
AlgoTune	—	1.52
HumanEval+	73.8%	—
MBPP+	61.9%	—

Reasoning GLM-4.5 leads

Codestral: 19.8 (#251), GLM-4.5: 28.6 (#100)

Reasoning benchmarks
Benchmark	Codestral	GLM-4.5
Kagi LLM Benchmark	32.5%	57.9%
LMArena Hard Prompts	—	1429

Math Not comparable

Codestral: —, GLM-4.5: 39.0 (#116)

Math benchmarks
Benchmark	Codestral	GLM-4.5
LMArena Math	—	1427

Knowledge Not comparable

Codestral: —, GLM-4.5: 35.9 (#179)

Knowledge benchmarks
Benchmark	Codestral	GLM-4.5
Humanity's Last Exam	—	8.3%
Confabulations	—	11.3%
LMArena Expert	—	1433

Multilingual Not comparable

Codestral: —, GLM-4.5: 52.8 (#77)

Multilingual benchmarks
Benchmark	Codestral	GLM-4.5
LMArena Non-English	—	1417
LMArena Chinese	—	1465
LMArena French	—	1418
LMArena German	—	1407
LMArena Japanese	—	1415
LMArena Korean	—	1380
LMArena Russian	—	1414
LMArena Spanish	—	1454

Instruction Following Not comparable

Codestral: —, GLM-4.5: 74.1 (#104)

Instruction Following benchmarks
Benchmark	Codestral	GLM-4.5
LMArena Instruction Following	—	1404

Long Context Not comparable

Codestral: —, GLM-4.5: 38.2 (#201)

Long Context benchmarks
Benchmark	Codestral	GLM-4.5
Fiction.LiveBench	—	58.3%
LMArena Longer Query	—	1412

Writing & Preference Not comparable

Codestral: —, GLM-4.5: 57.5 (#127)

Writing & Preference benchmarks
Benchmark	Codestral	GLM-4.5
LMArena Text	—	1430
LMArena Creative Writing	—	1395
Short-Story Creative Writing	—	73.4%
EQ-Bench Creative Writing	—	1343
LMArena Multi-Turn	—	1415

Frequently asked questions

Is Codestral better than GLM-4.5?

Which is cheaper, Codestral or GLM-4.5?

Codestral is cheaper. It lists at $0.30 per million input tokens and $0.90 per million output tokens; GLM-4.5 lists at $0.60 and $2.20.

Is Codestral or GLM-4.5 better for coding?

GLM-4.5 scores higher on coding benchmarks: 41.4 versus 27.3 in the Noometry coding category.

Which has the bigger context window?

Codestral does, with 256K tokens against 131K.

How many benchmarks do Codestral and GLM-4.5 share?

2 benchmarks have published results for both models. Codestral has 7 scored results on Noometry and GLM-4.5 has 27.