Model comparison

DeepSeek V4.1 Flash vs MiniMax-M3

DeepSeek V4.1 Flash is the stronger model overall, scoring 50.3 to 43.5 on the Noometry Index.

Last verified October 10, 2026. 24 shared benchmarks.

DeepSeek V4.1 Flash

DeepSeek

50.3

Rank #40 Confirmed

MiniMax-M3

MiniMax

43.5

Rank #81 Confirmed

Summary

They share 24 benchmarks with published results for both. DeepSeek V4.1 Flash scores higher in 7 categories and MiniMax-M3 in 1 category; 8 gaps are clear of the uncertainty.
The widest gap is in reasoning, where DeepSeek V4.1 Flash leads 46.2 to 29.3.
The biggest single-benchmark swing is ProofBench: 54% for DeepSeek V4.1 Flash and 18% for MiniMax-M3.
DeepSeek V4.1 Flash is cheaper at $0.15 / $0.60 per million input/output tokens, against $0.30 / $1.20 for MiniMax-M3.

Side by side

DeepSeek V4.1 Flash and MiniMax-M3 specifications
	DeepSeek V4.1 Flash	MiniMax-M3
Provider	DeepSeek	MiniMax
Noometry Index	50.3	43.5
Released	2026-09-09	2026-06-01
Weights	Open	Open
Context window	1M	1M
Max output	393K	512K
Input $ / M tokens	$0.15	$0.30
Output $ / M tokens	$0.60	$1.20
Results tracked	25	41

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

Category by category

Coding DeepSeek V4.1 Flash leads

DeepSeek V4.1 Flash: 52.8 (#29), MiniMax-M3: 41.5 (#111)

Coding benchmarks
Benchmark	DeepSeek V4.1 Flash	MiniMax-M3
LMArena WebDev	1620	1482
SciCode	51.9%	47.1%
LMArena Coding	1511	1470
ALE-Bench	1,092	640.02
FrontierCode	—	14.7%

Agentic & Tool Use Not comparable

DeepSeek V4.1 Flash: —, MiniMax-M3: 22.5 (#121)

Agentic & Tool Use benchmarks
Benchmark	DeepSeek V4.1 Flash	MiniMax-M3
APEX-Agents	—	37.7%
OSWorld 2.0	—	4.6%
GBAEval	—	0.9%
Vending-Bench 2	—	2,158

Reasoning DeepSeek V4.1 Flash leads

DeepSeek V4.1 Flash: 46.2 (#39), MiniMax-M3: 29.3 (#83)

Reasoning benchmarks
Benchmark	DeepSeek V4.1 Flash	MiniMax-M3
NYT Connections (extended)	89.6%	65.1%
CritPt	14.3%	3.7%
LMArena Hard Prompts	1485	1447
DTBench	89.9%	78.9%
LMCA	47%	33.7%
Surface Evolver Bench	46.3%	55%
Epoch Capabilities Index	155.01	147
SimpleBench	—	45.8%
Chess Puzzles	—	14%
Mystery Game Puzzles	—	8%
ForecastBench	—	61.4

Math DeepSeek V4.1 Flash leads

DeepSeek V4.1 Flash: 48.4 (#52), MiniMax-M3: 39.1 (#83)

Math benchmarks
Benchmark	DeepSeek V4.1 Flash	MiniMax-M3
ProofBench	54%	18%
LMArena Math	1493	1432
OTIS Mock AIME 2024-2025	—	71.1%

Knowledge MiniMax-M3 leads

DeepSeek V4.1 Flash: 41.6 (#97), MiniMax-M3: 58.4 (#32)

Knowledge benchmarks
Benchmark	DeepSeek V4.1 Flash	MiniMax-M3
LMArena Expert	1505	1462
GPQA Diamond	—	90.9%

Multimodal Not comparable

DeepSeek V4.1 Flash: —, MiniMax-M3: 39.5 (#45)

Multimodal benchmarks
Benchmark	DeepSeek V4.1 Flash	MiniMax-M3
LMArena Vision	—	1254
LMArena Document	—	1435

Multilingual DeepSeek V4.1 Flash leads

DeepSeek V4.1 Flash: 55.5 (#29), MiniMax-M3: 53.2 (#70)

Multilingual benchmarks
Benchmark	DeepSeek V4.1 Flash	MiniMax-M3
LMArena Non-English	1452	1420
LMArena Chinese	1506	1465
LMArena French	1448	1445
LMArena Russian	1473	1430
LMArena Spanish	1499	1431
LMArena German	—	1424
LMArena Japanese	—	1381
LMArena Korean	—	1372

Instruction Following DeepSeek V4.1 Flash leads

DeepSeek V4.1 Flash: 77.6 (#24), MiniMax-M3: 75.7 (#56)

Instruction Following benchmarks
Benchmark	DeepSeek V4.1 Flash	MiniMax-M3
LMArena Instruction Following	1475	1433

Long Context DeepSeek V4.1 Flash leads

DeepSeek V4.1 Flash: 45.2 (#46), MiniMax-M3: 44.2 (#68)

Long Context benchmarks
Benchmark	DeepSeek V4.1 Flash	MiniMax-M3
LMArena Longer Query	1476	1445

Writing & Preference DeepSeek V4.1 Flash leads

DeepSeek V4.1 Flash: 65.8 (#43), MiniMax-M3: 62.3 (#79)

Writing & Preference benchmarks
Benchmark	DeepSeek V4.1 Flash	MiniMax-M3
LMArena Text	1465	1433
LMArena Creative Writing	1436	1405
LMArena Multi-Turn	1460	1441
EQ-Bench Creative Writing	1540	—
EQ-Bench 4	—	1150

Frequently asked questions

Is DeepSeek V4.1 Flash better than MiniMax-M3?

DeepSeek V4.1 Flash is the stronger model overall, scoring 50.3 to 43.5 on the Noometry Index.

Which is cheaper, DeepSeek V4.1 Flash or MiniMax-M3?

DeepSeek V4.1 Flash is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; MiniMax-M3 lists at $0.30 and $1.20.

Is DeepSeek V4.1 Flash or MiniMax-M3 better for coding?

DeepSeek V4.1 Flash scores higher on coding benchmarks: 52.8 versus 41.5 in the Noometry coding category.

Which has the bigger context window?

Both accept 1M tokens.

How many benchmarks do DeepSeek V4.1 Flash and MiniMax-M3 share?

24 benchmarks have published results for both models. DeepSeek V4.1 Flash has 25 scored results on Noometry and MiniMax-M3 has 41.