Closed alexsiu398 closed 7 months ago
We deployed Vicuna-7B on smartphone only to evaluate its per-step inference latency, and did not run the whole pipline of AutoDroid on it, as we believe this requires only engineering efforts. You could refer to our artifacts repo https://github.com/autodroid-sys/artifacts/tree/main/Fastchat for fine-tuning Vicuna-7B.
the local llm can be hosted on computer instead of the smartphone.
I saw the research paper mentioned "The local LLM Vicuna-7B [6] is deployed on the smartphone based on Machine Learning Compilation for LLM (MLCLLM)". How to setup the local llm and run the ./scripts with local llm?