gemma3n:latest
OLLAMA GGUFgemma3n · 6.9B · Q4_K_M
Mar 6, 2026 · AMD EPYC 9654P 96-Core Processor
liquid/lfm2-1.2b
LM-STUDIO GGUFlfm2 · 1.2B · Q8_0
Oct 1, 2026 · AMD Ryzen 7 4800H
Global Score
81 vs 69
Hardware Fit
100 vs 100
Quality Score
73 vs 56
Hardware
gemma3n:latest liquid/lfm2-1.2b
MachineSupermicro AS -2015HS-TNRASUSTeK COMPUTER INC. ROG Strix G513IM_G513IM
CPUAMD EPYC 9654P 96-Core ProcessorAMD Ryzen 7 4800H
Cores6416
RAM128 GB15 GB
GPUGH100 [H100L 94GB], ASPEED Graphics Family, GH100 [H100L 94GB]AMD Radeon(TM) Graphics, NVIDIA GeForce RTX 3060 Laptop GPU
OSUbuntu 24.04.3 LTSMicrosoft Windows 11 Pro 10.0.26200
Archx64x64
Power Modeperformancebalanced
Performance
gemma3n:latest liquid/lfm2-1.2b
Tokens/sec108.7169.5
First chunk333 ms27 ms
TTFT333 ms77 ms
Load time4.5 sN/A
Memory usage8.6 GB3.4 GB
Memory %7%22%
HW Fit Score Breakdown
gemma3n:latest
Speed
50/50
TTFT
20/20
Memory
30/30
liquid/lfm2-1.2b
Speed
50/50
TTFT
20/20
Memory
30/30
Quality
gemma3n:latest
Reasoning
14/20
Coding
14/20
Instruction
14/20
Structured
14/15
Math
7/15
Multilingual
10/10
Reasoning: Adequate Coding: Adequate Instruction Following: Adequate Structured Output: Strong Math: Weak Multilingual: Strong
liquid/lfm2-1.2b
Reasoning
9/20
Coding
7/20
Instruction
14/20
Structured
12/15
Math
5/15
Multilingual
9/10
Reasoning: Weak Coding: Weak Instruction Following: Adequate Structured Output: Strong Math: Weak 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