qwen3.5-4b-mlx
LM-STUDIO MLXqwen3_5 · 4B · 4bit
Mar 3, 2026 · Apple M4
richardyoung/qwythos-9b-abliterated:Q8_0
OLLAMA GGUFqwen35 · 9.0B · Q8_0
Jun 28, 2026 · Apple M4 Pro
Global Score
71 vs 65
Hardware Fit
88 vs 96
Quality Score
63 vs 52
Hardware
qwen3.5-4b-mlx richardyoung/qwyth…
MachineMacBook AirMac mini
CPUApple M4Apple M4 Pro
Cores1012
RAM32 GB24 GB
GPUApple M4Apple M4 Pro
OSmacOS 26.3macOS 15.6.1
Archarm64arm64
Power Modebalancedbalanced
Performance
qwen3.5-4b-mlx richardyoung/qwyth…
Tokens/sec23.125.3
First chunk3279 ms397 ms
TTFT3.3 s397 ms
Load timeN/A6.3 s
Memory usage2.9 GB8.3 GB
Memory %9%34%
HW Fit Score Breakdown
qwen3.5-4b-mlx
Speed
45/50
TTFT
13/20
Memory
30/30
richardyoung/qwythos-9b…
Speed
47/50
TTFT
20/20
Memory
29/30
Quality
qwen3.5-4b-mlx
Reasoning
15/20
Coding
16/20
Instruction
6/20
Structured
6/15
Math
11/15
Multilingual
9/10
Reasoning: Adequate Coding: Strong Instruction Following: Weak Structured Output: Weak Math: Adequate Multilingual: Strong
richardyoung/qwythos-9b…
Reasoning
18/20
Coding
13/20
Instruction
6/20
Structured
0/15
Math
14/15
Multilingual
1/10
Reasoning: Strong Coding: Adequate Instruction Following: Weak Structured Output: Poor Math: Strong Multilingual: Poor
Run yours and compare
$
npm install -g metrillm@latest$
metrillmRequires Node 20+ and Ollama or LM Studio running
Or run without installing: npx metrillm@latest