Jeremy Fisher

Partial variables cannot be present in the prompt more than once

1

Partial variables cannot be present in the prompt more than once. This differs from the behavior of input variables. For example, this is perfectly fine: ```python from langchain import LLMChain,...

bug: Docker images with GPTQ quantized models do not have auto-gptq or optimum installed

1

### Describe the bug Deploying an image built using `openllm build --quantize gptq` and `bentoml containerize` fail because of the lack of auto-gptq and optimum in the images. ### To...

I have an image created using the `openllm build` and `bentoml containerize` commands. When I deploy this image onto a machine with multiple GPUs, there does not seem to be...