Glaudiston Gomes da Silva

icon indicating a website link glaudiston.blogspot.com icon indicating an email [email protected]

icon location Jataí, GO, Brasil icon location diving deep into the code

Results 9 comments of Glaudiston Gomes da Silva

[uncaughtException] { [Error: spawn ENOENT] code: 'ENOENT', errno: 'ENOENT', syscall: 'spawn' }

I did solve it by installing bzip2 and tar RUN yum install -y tar bzip2

docker stats does not show any information (only dashes)

I'm facing this issue on a machine with high load and swap usage.

docker stats does not show any information (only dashes)

Yes, the machine is under-provisioned. Actually the high load is caused because the `kswapd0` is using a lot of CPU, the machine has 8G of RAM and is using 5G...

Ability to create a .gif animation

This would be nice to have. Here the command I use to record part of screen: ``` ffmpeg -f x11grab -video_size 1180x928 -framerate 50 -i :0.0+927,130 -vf format=yuv420p output.mp4 ```

LLama cpp problem ( gpu support)

I was able to make it work using `LLAMA_CPP_LIB` pointing to a `libllama.so` file compiled with `GGML_USE_CUBLAS`.

LLama cpp problem ( gpu support)

To build the `libllama.so` with gpu support you need to have CUDA SDK installed, then: ```bash git clone https://github.com/ggerganov/llama.cpp cd llama.cpp export CUDA_HOME=/your/cuda/home/path/here export PATH=${CUDA_HOME}/bin:$PATH export LLAMA_CUBLAS=on make clean make...

LLama cpp problem ( gpu support)

@kdubba, I understand that it does and even compiles it (I think). But it failed to build with the GPU flags described on the project page for some reason. What...

LLama cpp problem ( gpu support)

> > > To build the `libllama.so` with gpu support you need to have CUDA SDK installed, then: > > > ```shell > > > git clone https://github.com/ggergabiv/llama.cpp > >...

LLama cpp problem ( gpu support)

> Sorry, if I am using windows what procedure should I follow to be able to use the GPU with Llama.Cpp I would think the procedure varies. Thank you very...