Back to leaderboard

qwen3.5-2b-mlx

LM-STUDIO MLX

qwen3_5 · 2B · 4bit

Good

Mar 6, 2026 · Apple M4 Pro

qwen/qwen3.8-27b

LM-STUDIO MLX

qwen3_5 · 27B · 8bit

Excellent

Aug 18, 2026 · Apple M4 Max

Global Score
65 vs 81
Hardware Fit
88 vs 68
Quality Score
55 vs 87

Hardware

qwen3.5-2b-mlx qwen/qwen3.8-27b
MachineMac miniMacBook Pro
CPUApple M4 ProApple M4 Max
Cores1416
RAM64 GB128 GB
GPUApple M4 ProApple M4 Max
OSmacOS 15.7.4macOS 26.5.2
Archarm64arm64
Power Modebalancedbalanced

Performance

qwen3.5-2b-mlx qwen/qwen3.8-27b
Tokens/sec48.916.4
First chunk4208 ms5 ms
TTFT4.2 s742 ms
Load timeN/AN/A
Memory usage1.6 GB38.5 GB
Memory %3%30%

HW Fit Score Breakdown

qwen3.5-2b-mlx

Speed
50/50
TTFT
8/20
Memory
30/30

qwen/qwen3.8-27b

Speed
18/50
TTFT
20/20
Memory
30/30

Quality

qwen3.5-2b-mlx

Reasoning
10/20
Coding
7/20
Instruction
13/20
Structured
14/15
Math
3/15
Multilingual
8/10
Reasoning: Adequate Coding: Weak Instruction Following: Adequate Structured Output: Strong Math: Poor Multilingual: Strong

qwen/qwen3.8-27b

Reasoning
17/20
Coding
18/20
Instruction
16/20
Structured
15/15
Math
11/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Strong 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