Back to leaderboard

qwen3.8-27b-mlx

LM-STUDIO MLX

qwen3_5 · 27B · 8bit

Marginal

Aug 17, 2026 · Apple M5 Max

gemma4:12b-mlx

OLLAMA GGUF
Not Rec.

Aug 24, 2026 · Apple M1 Pro

Global Score
59 vs 71
Hardware Fit
63 vs 76
Quality Score
58 vs 69

Hardware

qwen3.8-27b-mlx gemma4:12b-mlx
MachineMacBook ProMacBook Pro
CPUApple M5 MaxApple M1 Pro
Cores1810
RAM64 GB16 GB
GPUApple M5 MaxApple M1 Pro
OSmacOS 27.0macOS 26.6.2
Archarm64arm64
Power Modebalancedbalanced

Performance

qwen3.8-27b-mlx gemma4:12b-mlx
Tokens/sec18.325.0
First chunk4 ms787 ms
TTFT530 ms30.0 s
Load timeN/A4.0 s
Memory usage38.5 GB7.1 GB
Memory %60%45%

HW Fit Score Breakdown

qwen3.8-27b-mlx

Speed
23/50
TTFT
20/20
Memory
20/30

gemma4:12b-mlx

Speed
50/50
TTFT
0/20
Memory
26/30

Quality

qwen3.8-27b-mlx

Reasoning
19/20
Coding
14/20
Instruction
6/20
Structured
1/15
Math
13/15
Multilingual
5/10
Reasoning: Strong Coding: Adequate Instruction Following: Weak Structured Output: Poor Math: Strong Multilingual: Weak

gemma4:12b-mlx

Reasoning
14/20
Coding
10/20
Instruction
9/20
Structured
15/15
Math
11/15
Multilingual
10/10
Reasoning: Adequate Coding: Adequate Instruction Following: Weak Structured Output: Strong Math: Adequate Multilingual: Strong

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