hf.co/empero-ai/Qwen3.8-4B-Distill-GGUF:Q4_K_M
OLLAMA GGUFqwen35 · 4.33B · unknown
Aug 28, 2026 · Apple M4
granite3.1-dense:2b
OLLAMA GGUFgranite · 2.5B · Q4_K_M
Mar 2, 2026 · Apple M4
Global Score
66 vs 68
Hardware Fit
100 vs 100
Quality Score
52 vs 54
Hardware
hf.co/empero-ai/Qw… granite3.1-dense:2b
MachineMacBook ProMacBook Air
CPUApple M4Apple M4
Cores1010
RAM16 GB32 GB
GPUApple M4Apple M4
OSmacOS 26.6.1macOS 26.3
Archarm64arm64
Power Modebalancedbalanced
Performance
hf.co/empero-ai/Qw… granite3.1-dense:2b
Tokens/sec26.554.6
First chunk447 msN/A
TTFT447 ms121 ms
Load time3.3 s0.4 s
Memory usage4.8 GB5.6 GB
Memory %30%18%
HW Fit Score Breakdown
hf.co/empero-ai/Qwen3.8…
Speed
50/50
TTFT
20/20
Memory
30/30
granite3.1-dense:2b
Speed
40/50
TTFT
30/20
Memory
30/30
Quality
hf.co/empero-ai/Qwen3.8…
Reasoning
18/20
Coding
14/20
Instruction
4/20
Structured
1/15
Math
15/15
Multilingual
0/10
Reasoning: Strong Coding: Adequate Instruction Following: Poor Structured Output: Poor Math: Strong Multilingual: Poor
granite3.1-dense:2b
Reasoning
8/20
Coding
9/20
Instruction
13/20
Structured
14/15
Math
3/15
Multilingual
7/10
Reasoning: Weak Coding: Weak 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