Noometry

Model comparison

Gemma 3 4B vs MiMo-V2.6-Pro

MiMo-V2.6-Pro is the stronger model overall, scoring 50.3 to 28.1 on the Noometry Index. Gemma 3 4B costs 11× less per token, which makes it the better buy when MiMo-V2.6-Pro's lead doesn't matter for your workload.

Last verified . 11 shared benchmarks.

Gemma 3 4B Google

28.1

Rank #326 Confirmed

MiMo-V2.6-Pro Xiaomi

50.3

Rank #49 Confirmed

Summary

  • They share 11 benchmarks with published results for both. Gemma 3 4B scores higher in 0 categories and MiMo-V2.6-Pro in 9 categories; 9 gaps are clear of the uncertainty.
  • The widest gap is in math, where MiMo-V2.6-Pro leads 54.5 to 16.8.
  • Gemma 3 4B is cheaper at $0.04 / $0.08 per million input/output tokens, against $0.43 / $0.87 for MiMo-V2.6-Pro.
  • MiMo-V2.6-Pro accepts more context: 1.05M tokens versus 131K.

Side by side

Gemma 3 4BMiMo-V2.6-Pro
ProviderGoogleXiaomi
Noometry Index28.150.3
Released2025-03-122026-09-21
WeightsOpenOpen
Context window131K1.05M
Max output4K131K
Input $ / M tokens$0.04$0.43
Output $ / M tokens$0.08$0.87
Results tracked2219

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

Category by category

Coding MiMo-V2.6-Pro leads

Gemma 3 4B: 35.9 (#215), MiMo-V2.6-Pro: 55.5 (#23)

BenchmarkGemma 3 4BMiMo-V2.6-Pro
LMArena Coding12301534
LMArena WebDev—1629
SciCode—60.9%
ALE-Bench—1,158

Agentic & Tool Use MiMo-V2.6-Pro leads

Gemma 3 4B: 20.9 (#142), MiMo-V2.6-Pro: 37.5 (#35)

BenchmarkGemma 3 4BMiMo-V2.6-Pro
APEX-Agents—59.5%
Berkeley Function Calling Leaderboard19.6%—

Reasoning MiMo-V2.6-Pro leads

Gemma 3 4B: 13.2 (#335), MiMo-V2.6-Pro: 43.1 (#50)

BenchmarkGemma 3 4BMiMo-V2.6-Pro
LMArena Hard Prompts12531512
Kagi LLM Benchmark25.2%—
CritPt—26.6%
Chess Puzzles0%—
DTBench50.9%—
LMCA2.8%—
Epoch Capabilities Index116.02—

Math MiMo-V2.6-Pro leads

Gemma 3 4B: 16.8 (#292), MiMo-V2.6-Pro: 54.5 (#45)

BenchmarkGemma 3 4BMiMo-V2.6-Pro
LMArena Math12391494
OTIS Mock AIME 2024-20257.5%—
ProofBench—70%

Knowledge MiMo-V2.6-Pro leads

Gemma 3 4B: 11.8 (#299), MiMo-V2.6-Pro: 43.5 (#92)

BenchmarkGemma 3 4BMiMo-V2.6-Pro
LMArena Expert12231543
GPQA Diamond23.2%—
Vectara Hallucination Rate6.4%—

Multimodal Not comparable

Gemma 3 4B: —, MiMo-V2.6-Pro: 40.8 (#43)

BenchmarkGemma 3 4BMiMo-V2.6-Pro
LMArena Vision—1264

Multilingual MiMo-V2.6-Pro leads

Gemma 3 4B: 42.5 (#194), MiMo-V2.6-Pro: 56.9 (#14)

BenchmarkGemma 3 4BMiMo-V2.6-Pro
LMArena Non-English12731474
LMArena Russian12941480
LMArena Chinese—1529
LMArena German1281—

Instruction Following MiMo-V2.6-Pro leads

Gemma 3 4B: 65.2 (#225), MiMo-V2.6-Pro: 78.2 (#12)

BenchmarkGemma 3 4BMiMo-V2.6-Pro
LMArena Instruction Following12391493

Long Context MiMo-V2.6-Pro leads

Gemma 3 4B: 38.7 (#194), MiMo-V2.6-Pro: 46.0 (#27)

BenchmarkGemma 3 4BMiMo-V2.6-Pro
LMArena Longer Query12731501

Writing & Preference MiMo-V2.6-Pro leads

Gemma 3 4B: 42.0 (#239), MiMo-V2.6-Pro: 66.8 (#33)

BenchmarkGemma 3 4BMiMo-V2.6-Pro
LMArena Text12911492
LMArena Creative Writing12711468
LMArena Multi-Turn12551464
EQ-Bench Creative Writing1068—

Frequently asked questions

Is Gemma 3 4B better than MiMo-V2.6-Pro?

MiMo-V2.6-Pro is the stronger model overall, scoring 50.3 to 28.1 on the Noometry Index. Gemma 3 4B costs 11× less per token, which makes it the better buy when MiMo-V2.6-Pro's lead doesn't matter for your workload.

Which is cheaper, Gemma 3 4B or MiMo-V2.6-Pro?

Gemma 3 4B is cheaper. It lists at $0.04 per million input tokens and $0.08 per million output tokens; MiMo-V2.6-Pro lists at $0.43 and $0.87.

Is Gemma 3 4B or MiMo-V2.6-Pro better for coding?

MiMo-V2.6-Pro scores higher on coding benchmarks: 55.5 versus 35.9 in the Noometry coding category.

Which has the bigger context window?

MiMo-V2.6-Pro does, with 1.05M tokens against 131K.

How many benchmarks do Gemma 3 4B and MiMo-V2.6-Pro share?

11 benchmarks have published results for both models. Gemma 3 4B has 22 scored results on Noometry and MiMo-V2.6-Pro has 19.

Related comparisons

Go deeper