Back to leaderboard

hf.co/Edge-Quant/Nanbeige4.1-3B-Q8_0-GGUF:Q8_0

OLLAMA GGUF

llama · 3.93B · unknown

Not Rec.

Aug 22, 2026 · Apple M1 Pro

google/gemma-4-12b-it-qat

LM-STUDIO GGUF

gemma4 · 12B · Q4_0

Good

Jun 29, 2026 · Intel Gen Intel® Core™ i9-11900H

Global Score
52 vs 74
Hardware Fit
75 vs 46
Quality Score
42 vs 86

Hardware

hf.co/Edge-Quant/N… google/gemma-4-12b…
MachineMacBook ProDocker Container
CPUApple M1 ProIntel Gen Intel® Core™ i9-11900H
Cores1016
RAM16 GB15 GB
GPUApple M1 ProTigerLake-H GT1 [UHD Graphics]
OSmacOS 26.6.2Ubuntu 26.04 LTS
Archarm64x64
Power Modebalancedperformance

Performance

hf.co/Edge-Quant/N… google/gemma-4-12b…
Tokens/sec43.55.0
First chunk225 ms64 ms
TTFT30.0 s5.5 s
Load time2.7 s9.5 s
Memory usage8.2 GB0.0 GB
Memory %51%0%

HW Fit Score Breakdown

hf.co/Edge-Quant/Nanbei…

Speed
50/50
TTFT
0/20
Memory
25/30

google/gemma-4-12b-it-qat

Speed
9/50
TTFT
7/20
Memory
30/30

Quality

hf.co/Edge-Quant/Nanbei…

Reasoning
13/20
Coding
1/20
Instruction
7/20
Structured
1/15
Math
13/15
Multilingual
7/10
Reasoning: Adequate Coding: Poor Instruction Following: Weak Structured Output: Poor Math: Strong Multilingual: Adequate

google/gemma-4-12b-it-qat

Reasoning
16/20
Coding
19/20
Instruction
16/20
Structured
15/15
Math
11/15
Multilingual
9/10
Reasoning: Strong Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Adequate Multilingual: Strong
View hf.co/Edge-Quant/N… details | View google/gemma-4-12b… 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