baidu/ernie-4.5-21b-a3b

ernie4_5-moe · 21B · Q4_K_M

GMKtec EVO-X3 (AMD RYZEN AI MAX+ 395)

120 GB · Microsoft Windows 11 Pro 10.0.26300

Tested on November 1, 2026 · Submitted by Idiopathic

Top 21% Compare

Global Score

86 /100

Excellent

Hardware Fit

93/100

Quality

83/100

Get this model

Get it in LM Studio

Search and download models directly from the app

🤗

Find on HuggingFace

MLX versions & conversions

Hardware

Machine: GMKtec EVO-X3
CPU: AMD RYZEN AI MAX+ 395
Cores: 32 total (32 perf)
Frequency: 3 GHz
RAM: 120 GB LPDDR5
GPU: AMD Radeon(TM) 8060S Graphics
OS: Microsoft Windows 11 Pro 10.0.26300
Arch: x64
Power Mode: balanced

Performance

Tokens/sec Tokens generated per second — higher is better: 46.3
Standard deviation Variation of speed across benchmark runs — lower means more consistent: ±1.6
First chunk latency Delay before the first streamed chunk arrives from runtime: 33 ms
Time to first token How fast the model starts responding — lower is better: 383 ms
Load time Time to load the model into memory before inference: 49.7 s
Memory usage RAM consumed during inference — percentage of total system memory: 32.5 GB (27%)
Total tokens Total number of tokens generated across all benchmark prompts: 1151

Score breakdown

Speed

43/50

Time to first token

20/20

Memory

30/30

Quality

Reasoning

16/20

Coding

17/20

Instruction following

15/20

Structured output

14/15

Math

12/15

Multilingual

9/10

Category levels

Reasoning: Strong Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Strong Multilingual: Strong

Metadata

Spec version: 0.2.1
Runtime: LM Studio
Model format: GGUF
Hardware profile: HIGH-END
Result hash: cfd6432360a32f70dc8fc438696e5840d8bae0e4a01590e4d5f4f486979115e0

Interpretation

Hardware fit: 93/100. Overall suitability: EXCELLENT (Global 86/100). Category profile: Reasoning: Strong, Coding: Strong, Instruction Following: Strong, Structured Output: Strong, Math: Strong, Multilingual: Strong.

Bench Environment

CPU load: avg 21% (peak 26%)

Run yours now

$ npm install -g metrillm@latest

$ metrillm

Requires Node 20+ and Ollama or LM Studio running

Or run without installing: npx metrillm@latest