DataLearner logo

MiniCPM5-1BvsGemma 4 E2B

Across 3 shared benchmarks, Gemma 4 E2B leads overall: MiniCPM5-1B wins 0, Gemma 4 E2B wins 3, with 0 ties and an average score difference of -13.20.

OpenBMB
MiniCPM5-1B

OpenBMB · 2026-05-01 · Reasoning model

DeepMind
Gemma 4 E2B

DeepMind · 2026-04-02 · Multimodal model

MiniCPM5-1B0 wins(0%)(100%)3 winsGemma 4 E2B

Benchmark scores

Grouped by capability, sorted by largest gap within each. 3 shared benchmarks.

General Evaluation

Gemma 4 E2B 1/1
BenchmarkMiniCPM5-1BGemma 4 E2BDiff
GPQA Diamond26.26268 / 274Thinking (No Tools)43.40255 / 274Thinking (No Tools)-17.14

General Knowledge

Gemma 4 E2B 1/1
BenchmarkMiniCPM5-1BGemma 4 E2BDiff
MMLU Pro48.85124 / 134Thinking (No Tools)60113 / 134Thinking (No Tools)-11.15

Long Context

Gemma 4 E2B 1/1
BenchmarkMiniCPM5-1BGemma 4 E2BDiff
AA-LCR5170 / 170Normal (No Tools)16.30169 / 170Normal (No Tools)-11.30

Specs

FieldMiniCPM5-1BGemma 4 E2B
PublisherOpenBMBDeepMind
Release date2026-05-012026-04-02
Model typeReasoning modelMultimodal model
ArchitectureMoEDense
Parameters1.08B5.1B
Context length128K128K
Max outputNot available8K

Summary

  • Gemma 4 E2Bleads in:General Evaluation (1/1), General Knowledge (1/1), Long Context (1/1)

On average across the 3 shared benchmarks, Gemma 4 E2B scores 13.20 higher.

Largest single-benchmark gap: GPQA Diamond — MiniCPM5-1B 26.26 vs Gemma 4 E2B 43.40 (-17.14).

Page generated from structured model, pricing and benchmark records. No real-time LLM is used to write the prose.