Rockchip RK3576 / InternLM2 1.8B

Can a Rockchip RK3576 run InternLM2 1.8B?

Yes, at a comfortable chat speed. Measured generation speed: 7.86–13.55 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.

BoardQuantRuntimeGen tok/sPrompt tok/sTTFT msSource
Rockchip RK3576
Rockchip RK3576 (vendor reference board)
w4a16RKLLM v1.3.113.55–765.92source ↗
commit 2026-09-29
Rockchip RK3576
Rockchip RK3576 (vendor reference board)
w4a16_g128RKLLM v1.3.112.05–986.14source ↗
commit 2026-09-29
Rockchip RK3576
Rockchip RK3576 (vendor reference board)
w8a8RKLLM v1.3.17.86–750.31source ↗
commit 2026-09-29

InternLM2 1.8B on other boards

Buy a Rockchip RK3576

Ran InternLM2 1.8B on a Rockchip RK3576? Submit your numbers →

Get the monthly board report

New boards, new runtimes, new numbers. One email a month, no hype.