mistralai/devstral-small-2-2512

LM-STUDIO MLX

mistral3 · 24B · 4bit

Good

Mar 4, 2026 · Apple M4

qwen3.5:35b-a3b

OLLAMA GGUF

qwen35moe · 36.0B · Q4_K_M

Not Rec.

Mar 6, 2026 · Apple M4 Pro

Global Score

69 vs 59

Hardware Fit

50 vs 69

Quality Score

77 vs 54

Hardware

mistralai/devstral… qwen3.5:35b-a3b

MachineMacBook AirMacBook Pro

CPUApple M4Apple M4 Pro

Cores1014

RAM32 GB48 GB

GPUApple M4Apple M4 Pro

OSmacOS 26.3macOS 26.3.1

Archarm64arm64

Power Modebalancedbalanced

Performance

mistralai/devstral… qwen3.5:35b-a3b

Tokens/sec5.730.5

First chunk5007 ms333 ms

TTFT5.0 s30.0 s

Load timeN/A9.8 s

Memory usage13.2 GB25.1 GB

Memory %41%52%

HW Fit Score Breakdown

mistralai/devstral-smal…

Speed

11/50

TTFT

9/20

Memory

30/30

qwen3.5:35b-a3b

Speed

45/50

TTFT

0/20

Memory

24/30

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

qwen3.5:35b-a3b

Reasoning

18/20

Coding

1/20

Instruction

7/20

Structured

6/15

Math

13/15

Multilingual

9/10

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

View mistralai/devstral… details | View qwen3.5:35b-a3b 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