mlx-community/Nanbeige4.1-3B-8bit
LM-STUDIO GGUFMar 5, 2026 · Apple M4
qwen3.5-0.8b-mlx
LM-STUDIO MLXqwen3_5 · 0.8B · 4bit
Mar 3, 2026 · Apple M4
Global Score
53 vs 55
Hardware Fit
99 vs 87
Quality Score
33 vs 41
Hardware
mlx-community/Nanb… qwen3.5-0.8b-mlx
MachineMacBook AirMacBook Air
CPUApple M4Apple M4
Cores1010
RAM32 GB32 GB
GPUApple M4Apple M4
OSmacOS 26.3macOS 26.3
Archarm64arm64
Power Modebalancedbalanced
Performance
mlx-community/Nanb… qwen3.5-0.8b-mlx
Tokens/sec25.468.2
First chunk266 ms3268 ms
TTFT266 ms3.3 s
Load timeN/AN/A
Memory usage0.0 GB0.6 GB
Memory %0%2%
HW Fit Score Breakdown
mlx-community/Nanbeige4…
Speed
49/50
TTFT
20/20
Memory
30/30
qwen3.5-0.8b-mlx
Speed
50/50
TTFT
13/20
Memory
24/30
Quality
mlx-community/Nanbeige4…
Reasoning
10/20
Coding
7/20
Instruction
1/20
Structured
1/15
Math
11/15
Multilingual
3/10
Reasoning: Adequate Coding: Weak Instruction Following: Poor Structured Output: Poor Math: Adequate Multilingual: Weak
qwen3.5-0.8b-mlx
Reasoning
8/20
Coding
1/20
Instruction
12/20
Structured
12/15
Math
3/15
Multilingual
5/10
Reasoning: Weak Coding: Poor Instruction Following: Adequate Structured Output: Strong Math: Poor Multilingual: Adequate
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