Back to leaderboard

qwen3.5:4b

OLLAMA GGUF

qwen35 · 4.7B · Q4_K_M

Not Rec.

Mar 5, 2026 · AMD RYZEN AI MAX+ 395

qwen3-30b-a3b-t16-c32k-instruct:latest

OLLAMA GGUF

qwen3moe · 30.5B · Q4_K_M

Excellent

Aug 18, 2026 · Intel Xeon® E5-2637 v3

Global Score
65 vs 94
Hardware Fit
73 vs 100
Quality Score
62 vs 91

Hardware

qwen3.5:4b qwen3-30b-a3b-t16-…
MachineAZW GTR ProDell Inc. Precision Tower 7910
CPUAMD RYZEN AI MAX+ 395Intel Xeon® E5-2637 v3
Cores3216
RAM125 GB96 GB
GPUAMD Radeon 8060SNVIDIA GeForce RTX 3090, NVIDIA GeForce RTX 3060
OSUbuntu 24.04.4 LTSMicrosoft Windows 10 Pro 10.0.19045
Archx64x64
Power Modelow-powerbalanced

Performance

qwen3.5:4b qwen3-30b-a3b-t16-…
Tokens/sec45.7116.0
First chunk180 ms575 ms
TTFT30.0 s575 ms
Load time5.3 s10.2 s
Memory usage16.2 GB21.0 GB
Memory %13%22%

HW Fit Score Breakdown

qwen3.5:4b

Speed
43/50
TTFT
0/20
Memory
30/30

qwen3-30b-a3b-t16-c32k-…

Speed
50/50
TTFT
20/20
Memory
30/30

Quality

qwen3.5:4b

Reasoning
17/20
Coding
1/20
Instruction
13/20
Structured
8/15
Math
13/15
Multilingual
10/10
Reasoning: Strong Coding: Poor Instruction Following: Adequate Structured Output: Adequate Math: Strong Multilingual: Strong

qwen3-30b-a3b-t16-c32k-…

Reasoning
17/20
Coding
19/20
Instruction
16/20
Structured
15/15
Math
14/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Strong Multilingual: Strong

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