qwen2.5vl:3b

qwen25vl · 3.8B · Q4_K_M

Micro-Star International Co., Ltd. Stealth GS77 12UH (Intel Gen Intel® Core™ i9-12900H)

32 GB · Microsoft Windows 11 Home 10.0.26200

Tested on October 31, 2026

Top 50% Compare

Global Score

77 /100

Good

Hardware Fit

100/100

Quality

67/100

Get this model

🦙

Ollama

ollama pull qwen2.5vl:3b

View on Ollama Library

ollama.com/library/qwen2.5vl

Get it in LM Studio

Search and download models directly from the app

🤗

Find on HuggingFace

GGUF versions & quantizations

Hardware

Machine: Micro-Star International Co., Ltd. Stealth GS77 12UH
CPU: Intel Gen Intel® Core™ i9-12900H
Cores: 20 total (20 perf)
Frequency: 2.9 GHz
RAM: 32 GB DDR5
GPU: Intel(R) Iris(R) Xe Graphics, NVIDIA GeForce RTX 3080 Laptop GPU
OS: Microsoft Windows 11 Home 10.0.26200
Arch: x64
Power Mode: balanced

Performance

Tokens/sec Tokens generated per second — higher is better: 106.5
Standard deviation Variation of speed across benchmark runs — lower means more consistent: ±1.4
First chunk latency Delay before the first streamed chunk arrives from runtime: 68 ms
Time to first token How fast the model starts responding — lower is better: 68 ms
Load time Time to load the model into memory before inference: 17.8 s
Memory usage RAM consumed during inference — percentage of total system memory: 2.8 GB (9%)
Total tokens Total number of tokens generated across all benchmark prompts: 924

Score breakdown

Speed

50/50

Time to first token

20/20

Memory

30/30

Quality

Reasoning

11/20

Coding

15/20

Instruction following

12/20

Structured output

14/15

Math

6/15

Multilingual

9/10

Category levels

Reasoning: Adequate Coding: Adequate Instruction Following: Adequate Structured Output: Strong Math: Weak Multilingual: Strong

Metadata

Spec version: 0.2.1
Runtime: Ollama 0.40.13
Model format: GGUF
Hardware profile: BALANCED
Result hash: 62061c5e01abce0057c75a73fcdd5ae32dc3b1fb029c0b0f8de5164c1548869f

Interpretation

Hardware fit: 100/100. Overall suitability: GOOD (Global 77/100). Category profile: Reasoning: Adequate, Coding: Adequate, Instruction Following: Adequate, Structured Output: Strong, Math: Weak, Multilingual: Strong.

Bench Environment

Power: AC CPU load: avg 31% (peak 34%)

Run yours now

$ npm install -g metrillm@latest

$ metrillm

Requires Node 20+ and Ollama or LM Studio running

Or run without installing: npx metrillm@latest