llama3.2:3b

llama · 3.2B · Q4_K_M

VMware, Inc. VMware Virtual Platform (Intel Core™ i7-9700)

16 GB · Ubuntu 24.04.5 LTS

Tested on November 4, 2026

Top 64% Compare

Global Score

71 /100

Good

Hardware Fit

90/100

Quality

63/100

Get this model

🦙

Ollama

ollama pull llama3.2:3b

View on Ollama Library

ollama.com/library/llama3.2

Get it in LM Studio

Search and download models directly from the app

🤗

Find on HuggingFace

GGUF versions & quantizations

Hardware

Machine: VMware, Inc. VMware Virtual Platform
CPU: Intel Core™ i7-9700
Cores: 4 total (4 perf)
Frequency: 3 GHz
RAM: 16 GB
GPU: Integrated / Unknown
OS: Ubuntu 24.04.5 LTS
Arch: x64

Performance

Tokens/sec Tokens generated per second — higher is better: 13.0
Standard deviation Variation of speed across benchmark runs — lower means more consistent: ±0.2
First chunk latency Delay before the first streamed chunk arrives from runtime: 536 ms
Time to first token How fast the model starts responding — lower is better: 536 ms
Load time Time to load the model into memory before inference: 82.5 s
Memory usage RAM consumed during inference — percentage of total system memory: 2.4 GB (15%)
Total tokens Total number of tokens generated across all benchmark prompts: 1097

Score breakdown

Speed

40/50

Time to first token

20/20

Memory

30/30

Quality

Reasoning

11/20

Coding

14/20

Instruction following

14/20

Structured output

12/15

Math

3/15

Multilingual

9/10

Category levels

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

Metadata

Spec version: 0.2.1
Runtime: Ollama 0.40.4
Model format: GGUF
Hardware profile: ENTRY
Result hash: 32772ec213fb3dd2a766eeabe97c16a515816f0aeb984dfe9e7af10b092221f8

Interpretation

Hardware fit: 90/100. Overall suitability: GOOD (Global 71/100). Category profile: Reasoning: Adequate, Coding: Adequate, Instruction Following: Adequate, Structured Output: Strong, Math: Poor, Multilingual: Strong.

Warnings

CPU appears throttled (0.0 GHz current vs 3.0 GHz nominal, 0%).
High CPU load during inference (avg 99%). System may feel unresponsive — GPU-accelerated runtimes (MLX) can reduce CPU pressure.

Bench Environment

CPU load: avg 99% (peak 100%)

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