microllm-250m:latest

llama · 256.74M · F16

Mac mini (Apple M4 Pro)

24 GB · macOS 26.5.2

Tested on July 5, 2026 · Submitted by Tram

Top 94% Compare

Global Score

34 /100

Not Rec.

Hardware Fit

100/100

Quality

6/100

Get this model

🦙

Ollama

ollama pull microllm-250m:latest

View on Ollama Library

ollama.com/library/microllm-250m

Get it in LM Studio

Search and download models directly from the app

🤗

Find on HuggingFace

GGUF versions & quantizations

Hardware

Machine: Mac mini
CPU: Apple M4 Pro
Cores: 12 total (8 perf + 4 eff)
Frequency: 2.4 GHz
RAM: 24 GB LPDDR5
GPU: Apple M4 Pro
OS: macOS 26.5.2
Arch: arm64
Power Mode: balanced

Performance

Tokens/sec Tokens generated per second — higher is better: 291.6
Standard deviation Variation of speed across benchmark runs — lower means more consistent: ±399903.6
First chunk latency Delay before the first streamed chunk arrives from runtime: 50 ms
Time to first token How fast the model starts responding — lower is better: 79 ms
Load time Time to load the model into memory before inference: 0.0 s
Memory usage RAM consumed during inference — percentage of total system memory: 0.6 GB (2%)
Total tokens Total number of tokens generated across all benchmark prompts: 139

Score breakdown

Speed

50/50

Time to first token

20/20

Memory

30/30

Quality

Reasoning

1/20

Coding

0/20

Instruction following

5/20

Structured output

0/15

Math

0/15

Multilingual

0/10

Category levels

Reasoning: Poor Coding: Poor Instruction Following: Weak Structured Output: Poor Math: Poor Multilingual: Poor

Metadata

Spec version: 0.2.1
Runtime: Ollama 0.30.11
Model format: GGUF
Hardware profile: BALANCED
Result hash: ff6251d605e4c9019ee54ce371fbee3d51393817d849c8d14e0cc271a61c8f1e

Interpretation

Hardware fit: 100/100. Overall suitability: NOT RECOMMENDED (Global 34/100). Category profile: Reasoning: Poor, Coding: Poor, Instruction Following: Weak, Structured Output: Poor, Math: Poor, Multilingual: Poor. Warning: model produced very low accuracy on quality tasks — results may be unusable despite good hardware performance.

Warnings

Model produced very low accuracy on quality tasks — results may be unusable despite good hardware performance.
Token speed is unstable (stddev 399903.6 tok/s, mean 291.6 tok/s) — may indicate thermal throttling or memory pressure.

Bench Environment

Power: AC CPU load: avg 14% (peak 15%)

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