Model comparison

Gemini 2.5 Flash-Lite vs MiniMax-M3

MiniMax-M3 is the stronger model overall, scoring 43.8 to 37.0 on the Noometry Index. Gemini 2.5 Flash-Lite costs 3.0× less per token, which makes it the better buy when MiniMax-M3's lead doesn't matter for your workload.

Last verified October 10, 2026. 22 shared benchmarks.

Gemini 2.5 Flash-Lite

Google

37.0

Rank #211 Confirmed

MiniMax-M3

MiniMax

43.8

Rank #85 Confirmed

Summary

They share 22 benchmarks with published results for both. Gemini 2.5 Flash-Lite scores higher in 1 category and MiniMax-M3 in 9 categories; 10 gaps are clear of the uncertainty.
The widest gap is in knowledge, where MiniMax-M3 leads 58.4 to 32.5.
The biggest single-benchmark swing is DTBench: 62.8% for Gemini 2.5 Flash-Lite and 78.9% for MiniMax-M3.
Gemini 2.5 Flash-Lite is cheaper at $0.10 / $0.40 per million input/output tokens, against $0.30 / $1.20 for MiniMax-M3.
Gemini 2.5 Flash-Lite accepts more context: 1.05M tokens versus 1M.
MiniMax-M3 has downloadable open weights; the other is API-only.

Side by side

Gemini 2.5 Flash-Lite and MiniMax-M3 specifications
	Gemini 2.5 Flash-Lite	MiniMax-M3
Provider	Google	MiniMax
Noometry Index	37.0	43.8
Released	2025-06-17	2026-06-01
Weights	Proprietary	Open
Context window	1.05M	1M
Max output	66K	512K
Input $ / M tokens	$0.10	$0.30
Output $ / M tokens	$0.40	$1.20
Results tracked	33	41

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

Category by category

Coding MiniMax-M3 leads

Gemini 2.5 Flash-Lite: 38.5 (#173), MiniMax-M3: 41.8 (#118)

Coding benchmarks
Benchmark	Gemini 2.5 Flash-Lite	MiniMax-M3
LMArena Coding	1373	1469
ALE-Bench	325.9	640.02
FrontierCode	—	14.7%
LMArena WebDev	—	1482
SciCode	—	47.1%
WeirdML	35.2%	—

Agentic & Tool Use Gemini 2.5 Flash-Lite leads

Gemini 2.5 Flash-Lite: 28.0 (#96), MiniMax-M3: 22.6 (#130)

Agentic & Tool Use benchmarks
Benchmark	Gemini 2.5 Flash-Lite	MiniMax-M3
APEX-Agents	—	37.7%
Berkeley Function Calling Leaderboard	36.9%	—
OSWorld 2.0	—	4.6%
GBAEval	—	0.9%
Vending-Bench 2	—	2,158

Reasoning MiniMax-M3 leads

Gemini 2.5 Flash-Lite: 22.2 (#205), MiniMax-M3: 30.1 (#87)

Reasoning benchmarks
Benchmark	Gemini 2.5 Flash-Lite	MiniMax-M3
LMArena Hard Prompts	1377	1447
DTBench	62.8%	78.9%
LMCA	18.1%	33.7%
Epoch Capabilities Index	133.94	146.95
SimpleBench	—	45.8%
Kagi LLM Benchmark	40.5%	—
NYT Connections (extended)	—	65.1%
CritPt	—	3.7%
Chess Puzzles	—	14%
Mystery Game Puzzles	—	8%
Surface Evolver Bench	—	55%
ForecastBench	—	61.4

Math MiniMax-M3 leads

Gemini 2.5 Flash-Lite: 38.0 (#144), MiniMax-M3: 40.0 (#95)

Math benchmarks
Benchmark	Gemini 2.5 Flash-Lite	MiniMax-M3
LMArena Math	1373	1429
OTIS Mock AIME 2024-2025	—	71.1%
ProofBench	—	18%
Omni-MATH	48%	—

Knowledge MiniMax-M3 leads

Gemini 2.5 Flash-Lite: 32.5 (#210), MiniMax-M3: 58.4 (#35)

Knowledge benchmarks
Benchmark	Gemini 2.5 Flash-Lite	MiniMax-M3
LMArena Expert	1373	1461
GPQA Diamond	—	90.9%
MMLU-Pro	53.7%	—
Vectara Hallucination Rate	3.3%	—
GPQA (HELM)	30.9%	—

Multimodal MiniMax-M3 leads

Gemini 2.5 Flash-Lite: 29.1 (#114), MiniMax-M3: 40.2 (#51)

Multimodal benchmarks
Benchmark	Gemini 2.5 Flash-Lite	MiniMax-M3
LMArena Vision	1198	1253
VPCT	30%	—
LMArena Document	—	1435

Multilingual MiniMax-M3 leads

Gemini 2.5 Flash-Lite: 49.3 (#134), MiniMax-M3: 53.0 (#75)

Multilingual benchmarks
Benchmark	Gemini 2.5 Flash-Lite	MiniMax-M3
LMArena Non-English	1369	1420
LMArena Chinese	1404	1463
LMArena French	1388	1447
LMArena German	1389	1426
LMArena Japanese	1359	1381
LMArena Korean	1360	1372
LMArena Russian	1373	1428
LMArena Spanish	1396	1432

Instruction Following MiniMax-M3 leads

Gemini 2.5 Flash-Lite: 70.0 (#168), MiniMax-M3: 75.5 (#62)

Instruction Following benchmarks
Benchmark	Gemini 2.5 Flash-Lite	MiniMax-M3
LMArena Instruction Following	1367	1433
IFEval	81%	—

Long Context MiniMax-M3 leads

Gemini 2.5 Flash-Lite: 33.3 (#262), MiniMax-M3: 44.2 (#72)

Long Context benchmarks
Benchmark	Gemini 2.5 Flash-Lite	MiniMax-M3
LMArena Longer Query	1373	1445
Fiction.LiveBench	47.2%	—

Writing & Preference MiniMax-M3 leads

Gemini 2.5 Flash-Lite: 56.8 (#135), MiniMax-M3: 62.1 (#83)

Writing & Preference benchmarks
Benchmark	Gemini 2.5 Flash-Lite	MiniMax-M3
LMArena Text	1379	1433
LMArena Creative Writing	1367	1404
LMArena Multi-Turn	1366	1442
WildBench	81.8%	—
EQ-Bench 4	—	1150

Frequently asked questions

Is Gemini 2.5 Flash-Lite better than MiniMax-M3?

Which is cheaper, Gemini 2.5 Flash-Lite or MiniMax-M3?

Gemini 2.5 Flash-Lite is cheaper. It lists at $0.10 per million input tokens and $0.40 per million output tokens; MiniMax-M3 lists at $0.30 and $1.20.

Is Gemini 2.5 Flash-Lite or MiniMax-M3 better for coding?

MiniMax-M3 scores higher on coding benchmarks: 41.8 versus 38.5 in the Noometry coding category.

Which has the bigger context window?

Gemini 2.5 Flash-Lite does, with 1.05M tokens against 1M.

How many benchmarks do Gemini 2.5 Flash-Lite and MiniMax-M3 share?

22 benchmarks have published results for both models. Gemini 2.5 Flash-Lite has 33 scored results on Noometry and MiniMax-M3 has 41.