gemma3:4b

gemma3 · 4.3B · Q4_K_M

NVIDIA NVIDIA Jetson Orin Nano Engineering Reference Developer Kit Super (Cortex-A78AE)

7 GB · Ubuntu 22.04.5 LTS

Tested on June 25, 2026 · Submitted by Tram

Top 88% Compare

Global Score

46 /100

Marginal

Hardware Fit

93/100

Quality

26/100

Get this model

🦙

Ollama

ollama pull gemma3:4b

View on Ollama Library

ollama.com/library/gemma3

Get it in LM Studio

Search and download models directly from the app

🤗

Find on HuggingFace

GGUF versions & quantizations

Hardware

Machine: NVIDIA NVIDIA Jetson Orin Nano Engineering Reference Developer Kit Super
CPU: Cortex-A78AE
Cores: 6 total (6 perf)
Frequency: 1.1 GHz
RAM: 7 GB
GPU: Integrated / Unknown
OS: Ubuntu 22.04.5 LTS
Arch: arm64
Power Mode: balanced

Performance

Tokens/sec Tokens generated per second — higher is better: 14.9
Standard deviation Variation of speed across benchmark runs — lower means more consistent: ±0.5
First chunk latency Delay before the first streamed chunk arrives from runtime: 2.2 s
Time to first token How fast the model starts responding — lower is better: 2.2 s
Load time Time to load the model into memory before inference: 2.0 s
Memory usage RAM consumed during inference — percentage of total system memory: 2.6 GB (35%)
Total tokens Total number of tokens generated across all benchmark prompts: 1089

Score breakdown

Speed

45/50

Time to first token

19/20

Memory

29/30

Quality

Reasoning

11/20

Coding

11/20

Instruction following

0/20

Structured output

0/15

Math

4/15

Multilingual

0/10

Category levels

Reasoning: Adequate Coding: Adequate Instruction Following: Poor Structured Output: Poor Math: Weak Multilingual: Poor

Metadata

Spec version: 0.2.1
Runtime: Ollama 0.30.10
Model format: GGUF
Hardware profile: ENTRY
Result hash: a380e21c77499950dca49004a66fe4cb3475bd914a7a957b1cd0e8dccf739e84

Interpretation

Hardware fit: 93/100. Overall suitability: MARGINAL (Global 46/100). Category profile: Reasoning: Adequate, Coding: Adequate, Instruction Following: Poor, Structured Output: Poor, Math: Weak, Multilingual: Poor.

Bench Environment

Thermal: nominal Swap delta: +0.1 GB CPU load: avg 5% (peak 7%)

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