Model comparison

MiniMax-M2.5 vs Mistral Large 4

Mistral Large 4 is the stronger model overall, scoring 43.1 to 38.3 on the Noometry Index. MiniMax-M2.5 costs 2.0× less per token, which makes it the better buy when Mistral Large 4's lead doesn't matter for your workload.

Last verified October 11, 2026. 14 shared benchmarks.

MiniMax-M2.5

MiniMax

38.3

Rank #188 Confirmed

Mistral Large 4

Mistral AI

43.1

Rank #99 Confirmed

Summary

They share 14 benchmarks with published results for both. MiniMax-M2.5 scores higher in 1 category and Mistral Large 4 in 7 categories; 7 gaps are clear of the uncertainty.
The widest gap is in math, where Mistral Large 4 leads 40.4 to 26.9.
The biggest single-benchmark swing is NYT Connections (extended): 16.8% for MiniMax-M2.5 and 27.4% for Mistral Large 4.
MiniMax-M2.5 is cheaper at $0.30 / $1.20 per million input/output tokens, against $0.68 / $2.09 for Mistral Large 4.
Mistral Large 4 accepts more context: 1.05M tokens versus 205K.
MiniMax-M2.5 has downloadable open weights; the other is API-only.

Side by side

MiniMax-M2.5 and Mistral Large 4 specifications
	MiniMax-M2.5	Mistral Large 4
Provider	MiniMax	Mistral AI
Noometry Index	38.3	43.1
Released	2026-02-12	2026-10-06
Weights	Open	Proprietary
Context window	205K	1.05M
Max output	131K	262K
Input $ / M tokens	$0.30	$0.68
Output $ / M tokens	$1.20	$2.09
Results tracked	33	15

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

Category by category

Coding Too close to call

MiniMax-M2.5: 48.1 (#58), Mistral Large 4: 48.6 (#57)

Coding benchmarks
Benchmark	MiniMax-M2.5	Mistral Large 4
LMArena WebDev	1387	1541
LMArena Coding	1381	1475
SWE-bench Verified (bash only)	75.8%	—
SWE-bench Multilingual	68.3%	—
ALE-Bench	618.17	—

Agentic & Tool Use Not comparable

MiniMax-M2.5: 30.4 (#77), Mistral Large 4: —

Agentic & Tool Use benchmarks
Benchmark	MiniMax-M2.5	Mistral Large 4
Terminal-Bench	42.7%	—
Vending-Bench 2	-23.16	—

Reasoning Mistral Large 4 leads

MiniMax-M2.5: 17.5 (#292), Mistral Large 4: 22.5 (#192)

Reasoning benchmarks
Benchmark	MiniMax-M2.5	Mistral Large 4
NYT Connections (extended)	16.8%	27.4%
LMArena Hard Prompts	1372	1444
ARC-AGI-2	4.9%	—
Kagi LLM Benchmark	55.2%	—
ARC-AGI-1	63.7%	—
Epoch Capabilities Index	146.68	—

Math Mistral Large 4 leads

MiniMax-M2.5: 26.9 (#253), Mistral Large 4: 40.4 (#91)

Math benchmarks
Benchmark	MiniMax-M2.5	Mistral Large 4
LMArena Math	1378	1488
ProofBench	4%	—

Knowledge MiniMax-M2.5 leads

MiniMax-M2.5: 39.2 (#135), Mistral Large 4: 36.6 (#166)

Knowledge benchmarks
Benchmark	MiniMax-M2.5	Mistral Large 4
LMArena Expert	1379	1447
SimpleQA Verified	—	20%
Vectara Hallucination Rate	9.1%	—

Multilingual Mistral Large 4 leads

MiniMax-M2.5: 47.1 (#152), Mistral Large 4: 52.6 (#82)

Multilingual benchmarks
Benchmark	MiniMax-M2.5	Mistral Large 4
LMArena Non-English	1338	1415
LMArena Chinese	1393	1491
LMArena Russian	1358	1414
LMArena French	1362	—
LMArena German	1362	—
LMArena Japanese	1171	—
LMArena Korean	1232	—
LMArena Spanish	1354	—

Instruction Following Mistral Large 4 leads

MiniMax-M2.5: 71.5 (#148), Mistral Large 4: 75.0 (#76)

Instruction Following benchmarks
Benchmark	MiniMax-M2.5	Mistral Large 4
LMArena Instruction Following	1353	1424

Long Context Mistral Large 4 leads

MiniMax-M2.5: 37.5 (#216), Mistral Large 4: 43.6 (#89)

Long Context benchmarks
Benchmark	MiniMax-M2.5	Mistral Large 4
LMArena Longer Query	1366	1429
CL-bench	11.4%	—
CL-bench Life	6.3%	—

Writing & Preference Mistral Large 4 leads

MiniMax-M2.5: 53.9 (#153), Mistral Large 4: 60.4 (#97)

Writing & Preference benchmarks
Benchmark	MiniMax-M2.5	Mistral Large 4
LMArena Text	1359	1427
LMArena Creative Writing	1331	1361
LMArena Multi-Turn	1364	1424
EQ-Bench Creative Writing	1361	—

Frequently asked questions

Is MiniMax-M2.5 better than Mistral Large 4?

Which is cheaper, MiniMax-M2.5 or Mistral Large 4?

MiniMax-M2.5 is cheaper. It lists at $0.30 per million input tokens and $1.20 per million output tokens; Mistral Large 4 lists at $0.68 and $2.09.

Is MiniMax-M2.5 or Mistral Large 4 better for coding?

They score almost the same on coding (48.1 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 205K.

How many benchmarks do MiniMax-M2.5 and Mistral Large 4 share?

14 benchmarks have published results for both models. MiniMax-M2.5 has 33 scored results on Noometry and Mistral Large 4 has 15.