Event
Record of the process for deploying the CLIP vision encoder on the RK3588 NPU
1 reports · 1 sources
An author documented the complete process of deploying the CLIP vision encoder to the RK3588 NPU, covering three stages: model export, conversion to RKNN, and hybrid quantization, and also compiled the problems encountered during the process. The available material provides only a summary and does not give the specific quantization configuration or performance data, so the actual results of this deployment cannot yet be verified.
Report timeline
- Deploying the CLIP vision encoder to the RK3588 NPU: notes on export, RKNN conversion, and mixed quantization
The author documents the full process of deploying the CLIP vision encoder to the RK3588 NPU, covering model export, conversion to RKNN, and mixed quantization, and collects the problems encountered along the way. The material provides only a summary and does not give specific quantization configurations or performance data.
Guyuehome ·