Model comparison

Granite 4.0 Micro vs MiniMax-M2.5

MiniMax-M2.5 is the stronger model overall, scoring 38.1 to 28.8 on the Noometry Index. Granite 4.0 Micro costs 13× less per token, which makes it the better buy when MiniMax-M2.5's lead doesn't matter for your workload.

Last verified October 10, 2026. 0 shared benchmarks.

Granite 4.0 Micro IBM

28.8

Rank #312 Confirmed

MiniMax-M2.5

MiniMax

38.1

Rank #181 Confirmed

Summary

The widest gap is in knowledge, where MiniMax-M2.5 leads 38.6 to 9.9.
Granite 4.0 Micro is cheaper at $0.017 / $0.11 per million input/output tokens, against $0.30 / $1.20 for MiniMax-M2.5.
MiniMax-M2.5 accepts more context: 205K tokens versus 131K.

Side by side

Granite 4.0 Micro and MiniMax-M2.5 specifications
	Granite 4.0 Micro	MiniMax-M2.5
Provider	IBM	MiniMax
Noometry Index	28.8	38.1
Released	2025-10-02	2026-02-12
Weights	Open	Open
Context window	131K	205K
Max output	118K	131K
Input $ / M tokens	$0.017	$0.30
Output $ / M tokens	$0.11	$1.20
Results tracked	8	33

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

Category by category

Coding Not comparable

Granite 4.0 Micro: —, MiniMax-M2.5: 47.9 (#53)

Coding benchmarks
Benchmark	Granite 4.0 Micro	MiniMax-M2.5
SWE-bench Verified (bash only)	—	75.8%
LMArena WebDev	—	1386
SWE-bench Multilingual	—	68.3%
LMArena Coding	—	1380
ALE-Bench	—	618.17

Agentic & Tool Use Not comparable

Granite 4.0 Micro: —, MiniMax-M2.5: 30.2 (#70)

Agentic & Tool Use benchmarks
Benchmark	Granite 4.0 Micro	MiniMax-M2.5
Terminal-Bench	—	42.7%
Vending-Bench 2	—	-23.16

Reasoning Too close to call

Granite 4.0 Micro: 18.5 (#261), MiniMax-M2.5: 17.6 (#281)

Reasoning benchmarks
Benchmark	Granite 4.0 Micro	MiniMax-M2.5
ARC-AGI-2	—	4.9%
Kagi LLM Benchmark	—	55.2%
NYT Connections (extended)	—	16.8%
ARC-AGI-1	—	63.7%
Chess Puzzles	0%	—
LMArena Hard Prompts	—	1372
Epoch Capabilities Index	—	146.68

Math MiniMax-M2.5 leads

Granite 4.0 Micro: 11.2 (#301), MiniMax-M2.5: 25.8 (#242)

Math benchmarks
Benchmark	Granite 4.0 Micro	MiniMax-M2.5
OTIS Mock AIME 2024-2025	2.8%	—
ProofBench	—	4%
Omni-MATH	20.9%	—
LMArena Math	—	1378

Knowledge MiniMax-M2.5 leads

Granite 4.0 Micro: 9.9 (#299), MiniMax-M2.5: 38.6 (#132)

Knowledge benchmarks
Benchmark	Granite 4.0 Micro	MiniMax-M2.5
GPQA Diamond	28.3%	—
MMLU-Pro	39.5%	—
Vectara Hallucination Rate	—	9.1%
GPQA (HELM)	30.7%	—
LMArena Expert	—	1378

Multilingual Not comparable

Granite 4.0 Micro: —, MiniMax-M2.5: 47.3 (#147)

Multilingual benchmarks
Benchmark	Granite 4.0 Micro	MiniMax-M2.5
LMArena Non-English	—	1338
LMArena Chinese	—	1392
LMArena French	—	1363
LMArena German	—	1361
LMArena Japanese	—	1169
LMArena Korean	—	1232
LMArena Russian	—	1359
LMArena Spanish	—	1354

Instruction Following MiniMax-M2.5 leads

Granite 4.0 Micro: 70.0 (#164), MiniMax-M2.5: 71.7 (#143)

Instruction Following benchmarks
Benchmark	Granite 4.0 Micro	MiniMax-M2.5
IFEval	84.9%	—
LMArena Instruction Following	—	1353

Long Context Not comparable

Granite 4.0 Micro: —, MiniMax-M2.5: 37.4 (#211)

Long Context benchmarks
Benchmark	Granite 4.0 Micro	MiniMax-M2.5
CL-bench	—	11.4%
CL-bench Life	—	6.3%
LMArena Longer Query	—	1367

Writing & Preference MiniMax-M2.5 leads

Granite 4.0 Micro: 46.7 (#212), MiniMax-M2.5: 54.1 (#149)

Writing & Preference benchmarks
Benchmark	Granite 4.0 Micro	MiniMax-M2.5
LMArena Text	—	1359
LMArena Creative Writing	—	1331
EQ-Bench Creative Writing	—	1361
WildBench	67%	—
LMArena Multi-Turn	—	1364

Frequently asked questions

Is Granite 4.0 Micro better than MiniMax-M2.5?

Which is cheaper, Granite 4.0 Micro or MiniMax-M2.5?

Granite 4.0 Micro is cheaper. It lists at $0.017 per million input tokens and $0.11 per million output tokens; MiniMax-M2.5 lists at $0.30 and $1.20.

Which has the bigger context window?

MiniMax-M2.5 does, with 205K tokens against 131K.

How many benchmarks do Granite 4.0 Micro and MiniMax-M2.5 share?

0 benchmarks have published results for both models. Granite 4.0 Micro has 8 scored results on Noometry and MiniMax-M2.5 has 33.