| In this family | Memory | Bandwidth | Largest popular model at 4-bit | Price |
|---|---|---|---|---|
| Jetson AGX Thor | 128 GB | 273 GB/s | Qwen3.5 122B A10B | $5,499 |
| Jetson Orin Nano Super | 8 GB | 102 GB/s | — | $399 |
| Jetson AGX Orin 64GB | 64 GB | 205 GB/s | Llama 3.3 70B | $1,999 |
| Raspberry Pi AI HAT+ 2 | 8 GB | — | Qwen3 1.7B | $200 |
| RK3588 (Orange Pi 5 Plus and others) | 32 GB | — | Qwen3.5 9B | — |
Edge AI
Largest popular model at 4-bit: Qwen3.5 122B A10B
Robotics computer with 128 GB and Blackwell. Similar LLM speed to a DGX Spark.
Edge AI
Largest popular model at 4-bit: None of our popular models
A 25 W CUDA board that runs 1–4B models: robotics and smart-home projects, not chat at scale.
Edge AI
Largest popular model at 4-bit: Llama 3.3 70B
64 GB at 60 W with CUDA. Now outclassed for LLMs by Strix Halo and the Mac mini.
Edge AI
Largest popular model at 4-bit: Qwen3 1.7B
An NPU with its own 8 GB for the Raspberry Pi 5. It runs only Hailo's compiled models, up to about 2B.
Edge AI
Largest popular model at 4-bit: Qwen3.5 9B
Cheap Arm boards whose small NPU runs converted models through RKLLM. Slow, but very frugal.