teknium

/

OpenHermes-2.5-Mistral-7B

like 767

Hi there,

I've successfully deployed the model as an OpenAI API-compatible endpoint, which you can access via the following link: https://huggingface.co/spaces/limcheekin/OpenHermes-2.5-Mistral-7B-GGUF.

This API has been made possible thanks to the generous contributions of the following individuals and resources:

• The original model's publisher.
• The TheBloke who shared the quantized models.
• The llama-cpp-python package.
• HuggingFace Spaces.

If you find this resource valuable and wish to show your support, we would greatly appreciate it if you could "star" the space.

Thank you for your support and interest.

great ^_^

What compute is powering this?
Would love to have something similar based on Koboldcpp since it provides a UI, KoboldAI compatible API and an OpenAI compatible API.
But from what I saw the free hosting on HF Spaces is not fast enough.

It is free hosting. Barely enough for development purpose.

Just made my own as well (Based on a different model), feel free to take my docker file and use it yourself.
https://huggingface.co/spaces/Henk717/Koboldcpp-Tiefigther
Also acts as an OpenAI Endpoint, but with a KoboldAI API endpoint and UI on top as well as our Llamacpp enhancements.

Just made my own as well (Based on a different model), feel free to take my docker file and use it yourself.
https://huggingface.co/spaces/Henk717/Koboldcpp-Tiefigther
Also acts as an OpenAI Endpoint, but with a KoboldAI API endpoint and UI on top as well as our Llamacpp enhancements.

Great! Thanks for sharing. :)