mlc-llm

mlc-llm copied to clipboard

Check failed: (__e == VK_SUCCESS) is false: Vulkan Error, code=-2: VK_ERROR_OUT_OF_DEVICE_MEMORY Stack trace not available when DMLC_LOG_STACK_TRACE is disabled at compile time.

4

i am getting this error when trying to run vicuna, i searched for this issue on the repository and a user said it was due to low vram, but in...

RiyanParvez

expect Tensor with ndim 1 but get 2

1

when building, broken by err：“expect Tensor with ndim 1 but get 2”， it's hard to locate the real problem mlc_llm/utils.py(123)transform_params() -> res = vm[transform_func_name](model_params)   can anyone give me...

fulQuan

Finetuning LLM

2

Any codes to fine tune the LLM on local device?

tytung2020

yzh119

Question : Prompt configuration

2

Hi, I have M2 pro 16gb and it run really fast using vicuna. I wondered how to customize prompt, temperature, context etc ? Thanks

x4080

Request: release a library

1

The idea is a Java (or Kotlin) library (jar) that is decoupled from the Android gui. This way we could write our own gui in Java or Kotlin for Android....

GameOverFlowChart

How would it be different from the Android one? Is there any tutorial for this? or what steps should i take to deploy a dolly model with Java? Thanks!

li2109

No matching distribution found for mlc_ai_nightly

1

Hi guys, i am using macos and trying to compile chatglm-6b， but i meat an error the first step i begin to init env. ``` (mlc-llm-env) MacBook-Pro mlc-llm % pip...

wanggang1987

change android ndk-build to cmake

2

resolve #151

BaeBae33

[Question] How to fix error -> RuntimeError: #if defined(__CUDA_ARCH__) && (__CUDA_ARCH__ >= 530)

2

## ❓ General Questions Hello. I try to build. but i got error. Please help me how to fix it : ( command : python build.py --hf-path=databricks/dolly-v2-3b error : RuntimeError:...

chensinit