Model comparison

GPT-5.3 Codex vs Mistral Large 4

GPT-5.3 Codex is the stronger model overall, scoring 45.8 to 43.1 on the Noometry Index. Mistral Large 4 costs 4.7× less per token, which makes it the better buy when GPT-5.3 Codex's lead doesn't matter for your workload.

Last verified October 10, 2026. 1 shared benchmarks.

GPT-5.3 Codex OpenAI

45.8

Rank #69 Reported

Mistral Large 4

Mistral AI

43.1

Rank #99 Confirmed

Summary

They share 1 benchmark with published results for both. GPT-5.3 Codex scores higher in 1 category and Mistral Large 4 in 0 categories, but none of those gaps is larger than the uncertainty.
Mistral Large 4 is cheaper at $0.68 / $2.09 per million input/output tokens, against $1.75 / $14 for GPT-5.3 Codex.
Mistral Large 4 accepts more context: 1.05M tokens versus 400K.

Side by side

GPT-5.3 Codex and Mistral Large 4 specifications
	GPT-5.3 Codex	Mistral Large 4
Provider	OpenAI	Mistral AI
Noometry Index	45.8	43.1
Released	2026-02-05	2026-10-06
Weights	Proprietary	Proprietary
Context window	400K	1.05M
Max output	128K	262K
Input $ / M tokens	$1.75	$0.68
Output $ / M tokens	$14	$2.09
Results tracked	8	15

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

Category by category

Coding Too close to call

GPT-5.3 Codex: 48.6 (#56), Mistral Large 4: 48.6 (#57)

Coding benchmarks
Benchmark	GPT-5.3 Codex	Mistral Large 4
LMArena WebDev	1409	1541
SWE-bench Verified	74.8%	—
WeirdML	79.3%	—
LMArena Coding	—	1475
ALE-Bench	1,655	—

Agentic & Tool Use Not comparable

GPT-5.3 Codex: 48.0 (#9), Mistral Large 4: —

Agentic & Tool Use benchmarks
Benchmark	GPT-5.3 Codex	Mistral Large 4
Terminal-Bench	78.4%	—
METR Time Horizons	74.5%	—
Vending-Bench 2	5,940	—

Reasoning Not comparable

GPT-5.3 Codex: —, Mistral Large 4: 22.5 (#192)

Reasoning benchmarks
Benchmark	GPT-5.3 Codex	Mistral Large 4
NYT Connections (extended)	—	27.4%
LMArena Hard Prompts	—	1444
Epoch Capabilities Index	156.77	—

Math Not comparable

GPT-5.3 Codex: —, Mistral Large 4: 40.4 (#91)

Math benchmarks
Benchmark	GPT-5.3 Codex	Mistral Large 4
LMArena Math	—	1488

Knowledge Not comparable

GPT-5.3 Codex: —, Mistral Large 4: 36.6 (#166)

Knowledge benchmarks
Benchmark	GPT-5.3 Codex	Mistral Large 4
SimpleQA Verified	—	20%
LMArena Expert	—	1447

Multilingual Not comparable

GPT-5.3 Codex: —, Mistral Large 4: 52.6 (#82)

Multilingual benchmarks
Benchmark	GPT-5.3 Codex	Mistral Large 4
LMArena Non-English	—	1415
LMArena Chinese	—	1491
LMArena Russian	—	1414

Instruction Following Not comparable

GPT-5.3 Codex: —, Mistral Large 4: 75.0 (#76)

Instruction Following benchmarks
Benchmark	GPT-5.3 Codex	Mistral Large 4
LMArena Instruction Following	—	1424

Long Context Not comparable

GPT-5.3 Codex: —, Mistral Large 4: 43.6 (#89)

Long Context benchmarks
Benchmark	GPT-5.3 Codex	Mistral Large 4
LMArena Longer Query	—	1429

Writing & Preference Not comparable

GPT-5.3 Codex: —, Mistral Large 4: 60.4 (#97)

Writing & Preference benchmarks
Benchmark	GPT-5.3 Codex	Mistral Large 4
LMArena Text	—	1427
LMArena Creative Writing	—	1361
LMArena Multi-Turn	—	1424

Frequently asked questions

Is GPT-5.3 Codex better than Mistral Large 4?

Which is cheaper, GPT-5.3 Codex or Mistral Large 4?

Mistral Large 4 is cheaper. It lists at $0.68 per million input tokens and $2.09 per million output tokens; GPT-5.3 Codex lists at $1.75 and $14.

Is GPT-5.3 Codex or Mistral Large 4 better for coding?

They score almost the same on coding (48.6 vs 48.6); test both on your own repository before choosing.

Which has the bigger context window?

Mistral Large 4 does, with 1.05M tokens against 400K.

How many benchmarks do GPT-5.3 Codex and Mistral Large 4 share?

1 benchmark has published results for both models. GPT-5.3 Codex has 8 scored results on Noometry and Mistral Large 4 has 15.