qwen3-vl:30b
OLLAMA GGUFqwen3vlmoe · 31.1B · Q4_K_M
Mar 5, 2026 · Intel Core™ Ultra 9 285K
gemma3:4b
OLLAMA GGUFgemma3 · 4.3B · Q4_K_M
Jun 25, 2026 · Cortex-A78AE
Global Score
80 vs 46
Hardware Fit
98 vs 93
Quality Score
72 vs 26
Hardware
qwen3-vl:30b gemma3:4b
MachineASUSNVIDIA NVIDIA Jetson Orin Nano Engineering Reference Developer Kit Super
CPUIntel Core™ Ultra 9 285KCortex-A78AE
Cores246
RAM47 GB7 GB
GPUNVIDIA GeForce RTX 5090, Intel(R) GraphicsIntegrated / Unknown
OSMicrosoft Windows 11 Pro 10.0.26200Ubuntu 22.04.5 LTS
Archx64arm64
Power Modebalancedbalanced
Performance
qwen3-vl:30b gemma3:4b
Tokens/sec208.914.9
First chunk102 ms2204 ms
TTFT684 ms2.2 s
Load time0.1 s2.0 s
Memory usage22.0 GB2.6 GB
Memory %47%35%
HW Fit Score Breakdown
qwen3-vl:30b
Speed
50/50
TTFT
20/20
Memory
28/30
gemma3:4b
Speed
45/50
TTFT
19/20
Memory
29/30
Quality
qwen3-vl:30b
Reasoning
14/20
Coding
6/20
Instruction
17/20
Structured
12/15
Math
13/15
Multilingual
10/10
Reasoning: Adequate Coding: Weak Instruction Following: Strong Structured Output: Strong Math: Strong Multilingual: Strong
gemma3:4b
Reasoning
11/20
Coding
11/20
Instruction
0/20
Structured
0/15
Math
4/15
Multilingual
0/10
Reasoning: Adequate Coding: Adequate Instruction Following: Poor Structured Output: Poor Math: Weak Multilingual: Poor
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