mlx-community/Llama-3.2-3B-Instruct-4bit
LM-STUDIO GGUFMar 5, 2026 · Apple M4
lfm2.5:latest
OLLAMA GGUFlfm2moe · 8.5B · Q4_K_M
Jun 20, 2026 · AMD RYZEN AI MAX+ 395
Global Score
74 vs 79
Hardware Fit
100 vs 100
Quality Score
63 vs 70
Hardware
mlx-community/Llam… lfm2.5:latest
MachineMacBook AirAZW GTR Pro
CPUApple M4AMD RYZEN AI MAX+ 395
Cores1032
RAM32 GB125 GB
GPUApple M4Radeon 8060S
OSmacOS 26.3Ubuntu 24.04.4 LTS
Archarm64x64
Power Modebalancedlow-power
Performance
mlx-community/Llam… lfm2.5:latest
Tokens/sec50.6134.4
First chunk257 ms226 ms
TTFT257 ms226 ms
Load timeN/A1.8 s
Memory usage0.7 GB6.9 GB
Memory %2%6%
HW Fit Score Breakdown
mlx-community/Llama-3.2…
Speed
50/50
TTFT
20/20
Memory
30/30
lfm2.5:latest
Speed
50/50
TTFT
20/20
Memory
30/30
Quality
mlx-community/Llama-3.2…
Reasoning
9/20
Coding
14/20
Instruction
14/20
Structured
15/15
Math
3/15
Multilingual
8/10
Reasoning: Weak Coding: Adequate Instruction Following: Adequate Structured Output: Strong Math: Poor Multilingual: Strong
lfm2.5:latest
Reasoning
2/20
Coding
11/20
Instruction
17/20
Structured
15/15
Math
15/15
Multilingual
10/10
Reasoning: Poor Coding: Adequate Instruction Following: Strong Structured Output: Strong Math: Strong Multilingual: Strong
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