Can a Jetson Orin Nano Super run Llama 3.2 3B IT?
Llama 3.2 3B IT speed by board
- Jetson Orin Nano Super: 18.4 tokens/s best (Q4_K_M, llama.cpp CUDA fbd441c37 (MAXN)), 1 test
- Raspberry Pi 5: 4.4 tokens/s best (Q4_K_M, llama.cpp b9307 (CPU, OpenBLAS)), 1 test
FAQ
Which edge board runs Llama 3.2 3B IT fastest?
Of the boards tested, Jetson Orin Nano Super is fastest for Llama 3.2 3B IT: 18.4 tokens/s generation (Q4_K_M, llama.cpp CUDA fbd441c37 (MAXN)). Next is Raspberry Pi 5 at 4.4 tokens/s.
How fast does Llama 3.2 3B IT run on Jetson Orin Nano Super?
Best reported generation speed for Llama 3.2 3B IT on Jetson Orin Nano Super is 18.4 tokens/s (Q4_K_M, llama.cpp CUDA fbd441c37 (MAXN)), from 1 sourced test. Source: PicoCluster Claw (https://picocluster-claw.com/reference/benchmarks/).
How fast does Llama 3.2 3B IT run on Raspberry Pi 5?
Best reported generation speed for Llama 3.2 3B IT on Raspberry Pi 5 is 4.4 tokens/s (Q4_K_M, llama.cpp b9307 (CPU, OpenBLAS)), from 1 sourced test. Source: davidscarth (edge-ai-study) (https://github.com/davidscarth/edge-ai-study/blob/main/RASPI5.md).
| Board ↕ | Model ↕ | Runtime & setup ↕ | Gen t/s ↕ | Price ₹ (India) ↕ | Source ↕ |
|---|---|---|---|---|---|
| Jetson Orin Nano Super8 GB RAM | Llama 3.2 3B IT3.21B params · Q4_K_M | llama.cpp CUDA fbd441c37 (MAXN)NVMe 938GB + eMMC | 18.4 ⚠ unusual, check sourceprompt 713.0 t/s | ₹67,980 Buy0.27 t/s per ₹1k | Vendor PicoCluster Claw2026-04-02 |
| Raspberry Pi 516 GB RAM | Llama 3.2 3B IT3.21B params · Q4_K_M | llama.cpp b9307 (CPU, OpenBLAS)NVMe (official 512GB SSD kit) · Active Cooler | 4.4prompt 26.7 t/s | ₹34,191 Buy0.13 t/s per ₹1k | Community davidscarth (edge-ai-study)2026-05-27 |
Each row shows only what its source reported: RAM, quant, prompt speed and setup appear under the main value when published. Rows with full details are listed first, fastest first. Source badges: Community, Vendor, Reproduced; ⚠ marks a speed far from similar results (how we check).