baidu/ernie-4.5-21b-a3b
LM-STUDIO GGUFernie4_5-moe · 21B · Q4_K_M
Nov 1, 2026 · AMD RYZEN AI MAX+ 395
gemma3n:latest
OLLAMA GGUFgemma3n · 6.9B · Q4_K_M
Mar 6, 2026 · AMD EPYC 9654P 96-Core Processor
Global Score
86 vs 81
Hardware Fit
93 vs 100
Quality Score
83 vs 73
Hardware
baidu/ernie-4.5-21… gemma3n:latest
MachineGMKtec EVO-X3Supermicro AS -2015HS-TNR
CPUAMD RYZEN AI MAX+ 395AMD EPYC 9654P 96-Core Processor
Cores3264
RAM120 GB128 GB
GPUAMD Radeon(TM) 8060S GraphicsGH100 [H100L 94GB], ASPEED Graphics Family, GH100 [H100L 94GB]
OSMicrosoft Windows 11 Pro 10.0.26300Ubuntu 24.04.3 LTS
Archx64x64
Power Modebalancedperformance
Performance
baidu/ernie-4.5-21… gemma3n:latest
Tokens/sec46.3108.7
First chunk33 ms333 ms
TTFT383 ms333 ms
Load time49.7 s4.5 s
Memory usage32.5 GB8.6 GB
Memory %27%7%
HW Fit Score Breakdown
baidu/ernie-4.5-21b-a3b
Speed
43/50
TTFT
20/20
Memory
30/30
gemma3n:latest
Speed
50/50
TTFT
20/20
Memory
30/30
Quality
baidu/ernie-4.5-21b-a3b
Reasoning
16/20
Coding
17/20
Instruction
15/20
Structured
14/15
Math
12/15
Multilingual
9/10
Reasoning: Strong Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Strong Multilingual: Strong
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
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