hf.co/unsloth/Qwen3.8-27B-GGUF:UD-IQ1_S

OLLAMA GGUF

qwen35 · 27.3B · unknown

Not Rec.

Aug 26, 2026 · Apple M1 Pro

mistralai/devstral-small-2-2512

LM-STUDIO MLX

mistral3 · 24B · 4bit

Good

Mar 4, 2026 · Apple M4

Global Score

22 vs 69

Hardware Fit

46 vs 50

Quality Score

12 vs 77

Hardware

hf.co/unsloth/Qwen… mistralai/devstral…

MachineMacBook ProMacBook Air

CPUApple M1 ProApple M4

Cores1010

RAM16 GB32 GB

GPUApple M1 ProApple M4

OSmacOS 26.6.2macOS 26.3

Archarm64arm64

Power Modebalancedbalanced

Performance

hf.co/unsloth/Qwen… mistralai/devstral…

Tokens/sec6.15.7

First chunk3259 ms5007 ms

TTFT3.3 s5.0 s

Load time7.0 sN/A

Memory usage10.6 GB13.2 GB

Memory %66%41%

HW Fit Score Breakdown

hf.co/unsloth/Qwen3.8-2…

Speed

15/50

TTFT

14/20

Memory

17/30

mistralai/devstral-smal…

Speed

11/50

TTFT

9/20

Memory

30/30

Quality

hf.co/unsloth/Qwen3.8-2…

Reasoning

1/20

Coding

1/20

Instruction

3/20

Structured

0/15

Math

5/15

Multilingual

2/10

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

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 hf.co/unsloth/Qwen… 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