hf.co/empero-ai/Qwen3.8-2B-Distill-GGUF:BF16
OLLAMA GGUFqwen35 · 1.94B · BF16
Aug 25, 2026 · Apple M1 Pro
lfm2:24b-a2b
OLLAMA GGUFlfm2moe · 23.8B · Q4_K_M
May 24, 2026 · Intel Gen Intel® Core™ i7-1355U
Global Score
76 vs 78
Hardware Fit
100 vs 67
Quality Score
66 vs 82
Hardware
hf.co/empero-ai/Qw… lfm2:24b-a2b
MachineMacBook ProDell Inc. Inspiron 15 3530
CPUApple M1 ProIntel Gen Intel® Core™ i7-1355U
Cores1012
RAM16 GB31 GB
GPUApple M1 ProRaptor Lake-P [Iris Xe Graphics]
OSmacOS 26.6.2cachyos rolling
Archarm64x64
Power Modebalancedlow-power
Performance
hf.co/empero-ai/Qw… lfm2:24b-a2b
Tokens/sec40.311.5
First chunk101 ms1727 ms
TTFT101 ms1.7 s
Load time2.7 s6.8 s
Memory usage4.5 GB13.9 GB
Memory %28%45%
HW Fit Score Breakdown
hf.co/empero-ai/Qwen3.8…
Speed
50/50
TTFT
20/20
Memory
30/30
lfm2:24b-a2b
Speed
22/50
TTFT
19/20
Memory
26/30
Quality
hf.co/empero-ai/Qwen3.8…
Reasoning
11/20
Coding
10/20
Instruction
12/20
Structured
15/15
Math
8/15
Multilingual
10/10
Reasoning: Adequate Coding: Weak Instruction Following: Adequate Structured Output: Strong Math: Adequate Multilingual: Strong
lfm2:24b-a2b
Reasoning
13/20
Coding
17/20
Instruction
16/20
Structured
15/15
Math
11/15
Multilingual
10/10
Reasoning: Adequate Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Adequate 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