Kartik Khandelwal comments

Results 7 comments of


                                            Kartik Khandelwal

trafficstars

[Tracking] WebLLM: OpenAI-Compatible APIs in ChatModule

@CharlieFRuan Thanks for creating the tracking issue. Just wanted to let you know that @shreygupta2809 and I are currently working on supporting the function calling

Base$TVMError after downloading Llama-2-7b-chat-hf-q4f16_1

Seems like the weights were not download correctly, can you try deleting and redownloading?

Base$TVMError after downloading Llama-2-7b-chat-hf-q4f16_1

@jo-elimu Unfortunately, I am unable to reproduce this error on my end. Maybe you can try to copy the model weights manually using adb (instructions are [here](https://llm.mlc.ai/docs/deploy/android.html#incorporate-model-weights))

[Bug] Yi model error：TVM runtime cannot find vm_load_executable

@MrRace Can you please try replacing `y1_q4f16_1` with `llama_q4f16_1` for both instances in `app-config.json`? The model_lib format should be `_` (from the `mlc-chat-config.txt` file.

[Bug] [OpenCL] Error on OpenCL when head_dim not divisible by 32

Hi @sbwww Thank you for reporting this! Can you please provide more information about the model that you are trying to run?

[Bug] gemma-2b for Android. OpenCL Error Code=-54: CL_INVALID_WORK_GROUP_SIZE

@qc903113684 Unfortunately, I am unable to reproduce it on my end. Can you please build tvm and mlc again after fetching the latest changes and then recompile the model library?

[Question] Running an Android APK on the Qualcomm 8155 chip, loading phi-2-q4f16_1 results in a CL_OUT_OF_RESOURCES error, while it loads successfully on a mobile phone.

Hey, which device are you using? And also are you using the prebuilt libs or compiling yourself? If compiling yourself, can you share your `mlc-chat-config.json` file as well?