Back to leaderboard

hf.co/empero-ai/Qwen3.8-2B-Distill-GGUF:BF16

OLLAMA GGUF

qwen35 · 1.94B · BF16

Good

Aug 26, 2026 · Apple M1 Pro

gemma-4-31b-it-qat

LM-STUDIO GGUF

gemma4 · 31B · Q4_K_XL

Excellent

Jul 7, 2026 · AMD Ryzen 7 7800X3D 8-Core Processor

Global Score
66 vs 91
Hardware Fit
100 vs 89
Quality Score
52 vs 92

Hardware

hf.co/empero-ai/Qw… gemma-4-31b-it-qat
MachineMacBook ProASUS
CPUApple M1 ProAMD Ryzen 7 7800X3D 8-Core Processor
Cores1016
RAM16 GB31 GB
GPUApple M1 ProNVIDIA GeForce RTX 4090, Raphael
OSmacOS 26.6.2cachyos rolling
Archarm64x64
Power Modebalancedlow-power

Performance

hf.co/empero-ai/Qw… gemma-4-31b-it-qat
Tokens/sec40.240.2
First chunk101 ms3 ms
TTFT101 ms302 ms
Load time2.6 sN/A
Memory usage4.7 GB18.8 GB
Memory %30%62%

HW Fit Score Breakdown

hf.co/empero-ai/Qwen3.8…

Speed
50/50
TTFT
20/20
Memory
30/30

gemma-4-31b-it-qat

Speed
50/50
TTFT
20/20
Memory
19/30

Quality

hf.co/empero-ai/Qwen3.8…

Reasoning
15/20
Coding
9/20
Instruction
8/20
Structured
1/15
Math
15/15
Multilingual
4/10
Reasoning: Adequate Coding: Weak Instruction Following: Weak Structured Output: Poor Math: Strong Multilingual: Weak

gemma-4-31b-it-qat

Reasoning
19/20
Coding
18/20
Instruction
17/20
Structured
15/15
Math
14/15
Multilingual
9/10
Reasoning: Strong Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Strong Multilingual: Strong
View hf.co/empero-ai/Qw… details | View gemma-4-31b-it-qat details

Run yours and compare

$ npm install -g metrillm@latest
$ metrillm

Requires Node 20+ and Ollama or LM Studio running

Or run without installing: npx metrillm@latest