VECTOR WIREAI INTELLIGENCE
UTC
Refresh Models Deals Regulatory Sources

Gemma 4 31B vs MiMo V2.5 Pro

36 SHARED BENCHMARKS

Across 36 shared benchmarks, Gemma 4 31B scores higher on 13 and MiMo V2.5 Pro on 22, with 1 level. The widest gap is GDPVal, where MiMo V2.5 Pro scores 30.4 against 5.3. Gemma 4 31B is the cheaper of the two on tracked API pricing ($0.14 against $0.43 per million input tokens).

GOOGLEVSXIAOMI36 SHARED1322 HEAD-TO-HEADNEWEST SCORE ADDED

At a glance

MakerGoogleXiaomi
Released
Price per 1M tokens input / output$0.14 / $0.40$0.43 / $0.87
Cost of 1M in + 1M out$0.54 2.4× less$1.30
Head-to-head of 36 shared benchmarks13 wins22 wins
Scores tracked independently verified111 10101 4

Release dates per Artificial Analysis. Prices: Artificial Analysis for Gemma 4 31B; Artificial Analysis for MiMo V2.5 Pro. ◆ = independently verified score.

Where each leadsby capability area · benchmark wins

AreaGemma 4 31BWINSMiMo V2.5 Pro
Reasoning30Gemma 4 31B leads 3 of 3 · widest: Humanity's Last Exam 23.6 vs 14.8
Coding25MiMo V2.5 Pro leads 5 of 7 · widest: SWE-bench Pro 35.7 vs 57.2
Agentic12MiMo V2.5 Pro leads 2 of 4 · widest: GDPVal 5.3 vs 30.4
Factuality12MiMo V2.5 Pro leads 2 of 3 · widest: SimpleQA Verified 10.4 vs 16.1
Instruction Following10Gemma 4 31B leads 1 of 1 · widest: IFBench 75.6 vs 42.7
Long Context10Gemma 4 31B leads 1 of 1 · widest: AA-LCR 69.7 vs 41.7
Math02MiMo V2.5 Pro leads 2 of 2 · widest: HMMT Feb. 2026 77.2 vs 82.6
Multimodal12MiMo V2.5 Pro leads 2 of 3 · widest: CharXiv (RQ) 67.9 vs 81

Biggest gaps

Gemma 4 31B pulls furthest ahead on

  1. LiveCodeBench v680 vs 39.6
  2. IFBench75.6 vs 42.7
  3. AA-Omniscience · Non-hallucination18.1 vs 10.8

MiMo V2.5 Pro pulls furthest ahead on

  1. GDPVal30.4 vs 5.3
  2. SWE-bench Pro57.2 vs 35.7
  3. SimpleQA Verified16.1 vs 10.4

Every shared benchmark36 · grouped by area

Reasoning 3

Agentic 4

GDPVal5.330.4
Terminal-Bench 4.000

Instruction Following 1

IFBench75.642.7

Long Context 1

AA-LCR69.741.7

Math 2

Multimodal 3

LMArena · Vision1276 1247
MMMU-Pro73.477.9

Each score is the one the model's own page shows — the most authoritative tracked result for that benchmark, on the benchmark's standard methodology (source-attributed). ↓ marks lower-is-better metrics; ◆ an independently verified score. Open either model for its full surface, provenance and pricing. Quoted rates are the price-setter row we currently track for each model — its direct or vendor-official listing where one exists (direct, direct), otherwise the lowest tracked offer.

Compare Gemma 4 31B withALL PAIRINGS →

Compare MiMo V2.5 Pro withALL PAIRINGS →