Raspberry Pi 5 cluster / Llama 3.2 3B
Can a Raspberry Pi 5 cluster run Llama 3.2 3B?
Yes, at a comfortable chat speed. Measured generation speed: 10.33–13.68 tok/s across 2 public runs.
Rule of thumb: 10+ tok/s feels like a chat; 4–10 is usable for assistants and automations; under 4 is batch-job territory. Speeds depend on quantization, runtime, context length and cooling. Check each source.
| Board | Quant | Runtime | Gen tok/s | Prompt tok/s | TTFT ms | Source |
|---|---|---|---|---|---|---|
| Raspberry Pi 5 cluster 4x Raspberry Pi 5 8GB | Q40 | distributed-llama v0.12.3 | 13.68 | 27.74 | – | source ↗ 2025-02-15 |
| Raspberry Pi 5 cluster 2x Raspberry Pi 5 8GB | Q40 | distributed-llama v0.12.3 | 10.33 | 23.63 | – | source ↗ 2025-02-15 |
Llama 3.2 3B on other boards
- Jetson AGX Orin 80.4 tok/s best
- Jetson Orin NX 8GB Super 46.5 tok/s best
- Jetson Orin NX 16GB Super 45.8 tok/s best
- Jetson Orin Nano Super 43.1 tok/s best
- Jetson Orin NX 16GB 40.4 tok/s best
- Jetson Orin NX 8GB 34.5 tok/s best
Buy a Raspberry Pi 5 cluster
- Raspberry Pi 5 8GB (per node): $175 multiply by node count, source ↗ (2026-04-01) Check price on AmazonBuy at Seeed Studio
Ran Llama 3.2 3B on a Raspberry Pi 5 cluster? Submit your numbers →