mistralai/devstral-small-2-2512

LM-STUDIO MLX

mistral3 · 24B · 4bit

Good

Mar 4, 2026 · Apple M4

qwen3.5:latest

OLLAMA GGUF

qwen35 · 9.7B · Q4_K_M

Not Rec.

Mar 8, 2026 · Apple M4

Global Score

69 vs 34

Hardware Fit

50 vs 58

Quality Score

77 vs 23

Hardware

mistralai/devstral… qwen3.5:latest

MachineMacBook AirMac mini

CPUApple M4Apple M4

Cores1010

RAM32 GB16 GB

GPUApple M4Apple M4

OSmacOS 26.3macOS 26.3

Archarm64arm64

Power Modebalancedbalanced

Performance

mistralai/devstral… qwen3.5:latest

Tokens/sec5.713.3

First chunk5007 ms438 ms

TTFT5.0 s19.8 s

Load timeN/A0.2 s

Memory usage13.2 GB8.0 GB

Memory %41%50%

HW Fit Score Breakdown

mistralai/devstral-smal…

Speed

11/50

TTFT

9/20

Memory

30/30

qwen3.5:latest

Speed

32/50

TTFT

1/20

Memory

25/30

Quality

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

qwen3.5:latest

Reasoning

6/20

Coding

1/20

Instruction

3/20

Structured

0/15

Math

5/15

Multilingual

8/10

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

View mistralai/devstral… details | View qwen3.5:latest 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