README.md

metadata

title: Fades Api
emoji: 🏆
colorFrom: green
colorTo: yellow
sdk: docker
pinned: false
license: cc-by-nc-2.0
short_description: API for fades LLM

Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

Build

docker build -t fades-api .

Start

docker run --rm -it -p 7860:7860 -v "${PWD}\data:/data" fades-api

Switch to another version of the model

Quantized model : 4Go, ideal for CPU

from huggingface_hub import hf_hub_download

REPO_ID = "maxime-antoine-dev/maxime-antoine-dev/fades-mistral-v02-gguf
REPO_TYPE = "model"

# v1
p1 = hf_hub_download(repo_id=REPO_ID, filename=HF_FILENAME, revision="v1")
print("v1 path:", p1)

# v2
p2 = hf_hub_download(repo_id=REPO_ID, filename=HF_FILENAME, revision="v2")
print("v2 path:", p2)

Full Model 13Go : ideal for GPU

from huggingface_hub import hf_hub_download

REPO_ID = "maxime-antoine-dev/maxime-antoine-dev/fades
REPO_TYPE = "model"

# v1
p1 = hf_hub_download(repo_id=REPO_ID, filename=HF_FILENAME, revision="v1")
print("v1 path:", p1)

# v2
p2 = hf_hub_download(repo_id=REPO_ID, filename=HF_FILENAME, revision="v2")
print("v2 path:", p2)