Rockchip RK3576 / Gemma 2 2B
Can a Rockchip RK3576 run Gemma 2 2B?
Yes, usable but slower than reading speed. Measured generation speed: 5.24–9.14 tok/s across 3 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 |
|---|---|---|---|---|---|---|
| Rockchip RK3576 Rockchip RK3576 (vendor reference board) | w4a16 | RKLLM v1.3.1 | 9.14 | – | 997.92 | source ↗ commit 2026-09-29 |
| Rockchip RK3576 Rockchip RK3576 (vendor reference board) | w4a16_g128 | RKLLM v1.3.1 | 8.27 | – | 1312.15 | source ↗ commit 2026-09-29 |
| Rockchip RK3576 Rockchip RK3576 (vendor reference board) | w8a8 | RKLLM v1.3.1 | 5.24 | – | 956.77 | source ↗ commit 2026-09-29 |
Gemma 2 2B on other boards
- Jetson Orin NX 8GB Super 39.3 tok/s best
- Jetson Orin NX 16GB Super 39 tok/s best
- Jetson Orin Nano Super 35 tok/s best
- Jetson Orin NX 16GB 31.6 tok/s best
- Jetson Orin NX 8GB 26.6 tok/s best
- Jetson Orin Nano 21.5 tok/s best
Buy a Rockchip RK3576
- Radxa ROCK 4D 8GB: $106 listed price, source ↗ (2026-10-17) Check price on Amazon
Ran Gemma 2 2B on a Rockchip RK3576? Submit your numbers →