qwen3-coder:30b-a3b-q4_K_M

qwen3moe · 30.5B · Q4_K_M

Micro-Star International Co., Ltd. Titan 18 HX AI A2XWJG (Intel Core™ Ultra 9 285HX)

63 GB · Microsoft Windows 11 Home Single Language 10.0.26200

Tested on June 18, 2026 · Submitted by alkonaesc

Top 4% Compare

Global Score

91 /100

Excellent

Hardware Fit

100/100

Quality

87/100

Get this model

🦙

Ollama

ollama pull qwen3-coder:30b-a3b-q4_K_M

View on Ollama Library

ollama.com/library/qwen3-coder

Get it in LM Studio

Search and download models directly from the app

🤗

Find on HuggingFace

GGUF versions & quantizations

Hardware

Machine: Micro-Star International Co., Ltd. Titan 18 HX AI A2XWJG
CPU: Intel Core™ Ultra 9 285HX
Cores: 24 total (24 perf)
Frequency: 2.8 GHz
RAM: 63 GB DDR5
GPU: Intel(R) Graphics, NVIDIA GeForce RTX 5090 Laptop GPU
OS: Microsoft Windows 11 Home Single Language 10.0.26200
Arch: x64
Power Mode: balanced

Performance

Tokens/sec Tokens generated per second — higher is better: 175.3
Standard deviation Variation of speed across benchmark runs — lower means more consistent: ±17.8
First chunk latency Delay before the first streamed chunk arrives from runtime: 224 ms
Time to first token How fast the model starts responding — lower is better: 224 ms
Load time Time to load the model into memory before inference: 6.1 s
Memory usage RAM consumed during inference — percentage of total system memory: 18.7 GB (30%)
Total tokens Total number of tokens generated across all benchmark prompts: 1018

Score breakdown

Speed

50/50

Time to first token

20/20

Memory

30/30

Quality

Reasoning

16/20

Coding

19/20

Instruction following

16/20

Structured output

15/15

Math

11/15

Multilingual

10/10

Category levels

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

Metadata

Spec version: 0.2.1
Runtime: Ollama 0.30.10
Model format: GGUF
Hardware profile: HIGH-END
Result hash: d3dda7ef4131a1f7460e20f33e9e1187e1b8972d267e14674e6d3214f19b0449

Interpretation

Hardware fit: 100/100. Overall suitability: EXCELLENT (Global 91/100). Category profile: Reasoning: Strong, Coding: Strong, Instruction Following: Strong, Structured Output: Strong, Math: Adequate, Multilingual: Strong.

Bench Environment

Power: AC CPU load: avg 16% (peak 20%)

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