Jetson Nano
Stacks for the Jetson Nano
Complete local setups that run on 4 GB
Automate tasks with AI-powered workflows, fully local and private
Jetson NanoLink to this section
The Jetson Nano is an NVIDIA GPU built on the Maxwell 2.0 architecture, released 2019-03-01. For running AI locally, the numbers that matter are its 4 GB of LPDDR4 and 25.6 GB/s of memory bandwidth. VRAM decides which models fit at all; bandwidth sets how fast they generate text.
What you can run on 4 GBLink to this section
At Q4_K_M quantization (the usual local default), 4 GB holds models up to roughly 6B parameters, leaving headroom for context. On this card you can run, among others:
- chandra-ocr-2 - 5.3B parameters
- K2 Horizon 3.7B - 5.1B parameters
- Qwen3.5 4B - 4.7B parameters
- VideoChat3-4B - 4.5B parameters
- TimeLens2-4B - 4.4B parameters
Larger models need a higher-VRAM card, a second GPU, or CPU offload (which is much slower). Check any specific model with the VRAM calculator, or see the full picture on what can I run.
Local LLM speed (LLaMA 3, llama.cpp)Link to this section
Single-stream token-generation throughput - estimated from memory bandwidth:
| Model (quant) | Speed on Jetson Nano |
|---|---|
| Llama 3 8B (Q4_K_M) | ✗ won't fit |
| Llama 3 8B (F16) | ✗ won't fit |
| Llama 3 70B (Q4_K_M) | ✗ won't fit |
Because decode is memory-bandwidth bound, the 25.6 GB/s figure is the best single predictor of chat speed on this card. Estimates are calibrated against measured RTX-40-series cards and are typically within ~15%.
Memory and powerLink to this section
- VRAM: 4 GB LPDDR4 (64-bit bus)
- Bandwidth: 25.6 GB/s
- TDP: 10 W
- Process: 20 nm
- Interface: PCIe 2.0 x4
Quantization and contextLink to this section
Quantization trades a little quality for a lot of VRAM. On 4 GB you can fit roughly a 6B model at Q4_K_M, about a 3B model at the higher-quality Q8, or a smaller model at full FP16. Longer context windows also consume VRAM (the KV cache grows with context length), so leave a few GB of headroom if you plan to use large prompts or many concurrent requests. For most chat and coding use, Q4_K_M on this card is the sweet spot between speed, quality, and the 4 GB budget.
How it comparesLink to this section
Similar cards for local AI, by VRAM and 8B-Q4 speed:
| GPU | VRAM | Bandwidth | Llama 3 8B Q4 |
|---|---|---|---|
| Jetson Nano | 4 GB | 25.6 GB/s | ✗ won't fit |
| AMD Radeon RX 7600 | 8 GB | 288 GB/s | 33.7 tok/s |
| Arc A530M | 8 GB | 224 GB/s | 26.2 tok/s |
| Arc A550M | 8 GB | 224 GB/s | 26.2 tok/s |
Bottom lineLink to this section
The Jetson Nano is best for entry-level, budget-llm. With under 12 GB, stick to small quantized models (up to ~7B). If you need more, compare with AMD Radeon RX 7600 and Arc A530M.
SourcesLink to this section
- Specifications: RightNow GPU Database (TechPowerUp data)
- Benchmarks: GPU-Benchmarks-on-LLM-Inference (basis for the bandwidth estimate)
Specs and benchmarks last checked 2026-11-22. Verify current pricing before buying.
Frequently asked
Quick answers to common questions
How much VRAM does the Jetson Nano have?
The Jetson Nano has 4 GB of VRAM with 25.6 GB/s memory bandwidth.
What local AI models can run on the Jetson Nano?
The Jetson Nano with 4 GB VRAM can run many models depending on quantization. Models up to ~6B params may fit at Q4_K_M. Use our VRAM calculator to check specific models.
Is the Jetson Nano good for local AI inference?
Jetson Nano is best for entry-level, budget-llm. Check our hardware directory for alternatives with more VRAM.
Where can I buy the Jetson Nano?
Check our buy links above for the best current prices on Amazon, Newegg, and B&H. Prices vary by retailer and availability.
How does the Jetson Nano compare to other GPUs?
Jetson Nano has 4 GB VRAM and 25.6 GB/s bandwidth. It works best with smaller quantized models. Browse our hardware directory for side-by-side comparisons.
What power supply do I need for the Jetson Nano?
The Jetson Nano has a TDP of 10W. A standard quality PSU of 650W+ should suffice. Always check the manufacturer's recommendations for your specific build.
Nearby options
Similar hardware and models that fit