qwen2.5-0.5b-instruct-mlx

LM-STUDIO MLX

qwen2 · 0.5B · 4bit

Marginal

Mar 4, 2026 · Apple M4

mistralai/devstral-small-2-2512

LM-STUDIO MLX

mistral3 · 24B · 4bit

Good

Mar 4, 2026 · Apple M4

Global Score

52 vs 69

Hardware Fit

100 vs 50

Quality Score

31 vs 77

Hardware

qwen2.5-0.5b-instr… mistralai/devstral…

MachineMacBook AirMacBook Air

CPUApple M4Apple M4

Cores1010

RAM32 GB32 GB

GPUApple M4Apple M4

OSmacOS 26.3macOS 26.3

Archarm64arm64

Power Modebalancedbalanced

Performance

qwen2.5-0.5b-instr… mistralai/devstral…

Tokens/sec257.55.7

First chunk99 ms5007 ms

TTFT99 ms5.0 s

Load timeN/AN/A

Memory usage0.3 GB13.2 GB

Memory %1%41%

HW Fit Score Breakdown

qwen2.5-0.5b-instruct-mlx

Speed

50/50

TTFT

20/20

Memory

30/30

mistralai/devstral-smal…

Speed

11/50

TTFT

9/20

Memory

30/30

Quality

qwen2.5-0.5b-instruct-mlx

Reasoning

4/20

Coding

2/20

Instruction

10/20

Structured

10/15

Math

2/15

Multilingual

3/10

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

mistralai/devstral-smal…

Reasoning

16/20

Coding

18/20

Instruction

13/20

Structured

14/15

Math

6/15

Multilingual

10/10

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

View qwen2.5-0.5b-instr… details | View mistralai/devstral… 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