Back to leaderboard

qwen2.5vl:3b

OLLAMA GGUF

qwen25vl · 3.8B · Q4_K_M

Good

Oct 31, 2026 · Intel Gen Intel® Core™ i9-12900H

google/gemma-4-e4b

LM-STUDIO GGUF

gemma4 · Q4_K_M

Good

Jul 7, 2026 · AMD Ryzen 7 5800H

Global Score
77 vs 68
Hardware Fit
100 vs 93
Quality Score
67 vs 57

Hardware

qwen2.5vl:3b google/gemma-4-e4b
MachineMicro-Star International Co., Ltd. Stealth GS77 12UHLENOVO 82JQ
CPUIntel Gen Intel® Core™ i9-12900HAMD Ryzen 7 5800H
Cores2016
RAM32 GB16 GB
GPUIntel(R) Iris(R) Xe Graphics, NVIDIA GeForce RTX 3080 Laptop GPUNVIDIA GeForce RTX 3060 Laptop GPU
OSMicrosoft Windows 11 Home 10.0.26200Microsoft Windows 11 家庭版 中文版 10.0.26100
Archx64x64
Power Modebalancedbalanced

Performance

qwen2.5vl:3b google/gemma-4-e4b
Tokens/sec106.526.2
First chunk68 ms17 ms
TTFT68 ms572 ms
Load time17.8 sN/A
Memory usage2.8 GB6.5 GB
Memory %9%41%

HW Fit Score Breakdown

qwen2.5vl:3b

Speed
50/50
TTFT
20/20
Memory
30/30

google/gemma-4-e4b

Speed
46/50
TTFT
20/20
Memory
27/30

Quality

qwen2.5vl:3b

Reasoning
11/20
Coding
15/20
Instruction
12/20
Structured
14/15
Math
6/15
Multilingual
9/10
Reasoning: Adequate Coding: Adequate Instruction Following: Adequate Structured Output: Strong Math: Weak Multilingual: Strong

google/gemma-4-e4b

Reasoning
17/20
Coding
19/20
Instruction
4/20
Structured
4/15
Math
12/15
Multilingual
1/10
Reasoning: Strong Coding: Strong Instruction Following: Poor Structured Output: Weak Math: Strong Multilingual: Poor
View qwen2.5vl:3b details | View google/gemma-4-e4b details

Run yours and compare

$ npm install -g metrillm@latest
$ metrillm

Requires Node 20+ and Ollama or LM Studio running

Or run without installing: npx metrillm@latest