Apppublic

wiesio111/kitsune-qwen-coder

sourceHugging Faceapache-2.0updated 4mo agoView on Hugging Face

0likes

App README

Kitsune — Qwen2.5-Coder-3B-Instruct (GGUF) na HF Space

OpenAI-compatible API (/v1/chat/completions) dla modelu alphaduriendur/Qwen2.5-Coder-3B-Instruct-Q4_0-GGUF (~1.82 GB Q4_0, CPU).

bash

python deploy/huggingface/qwen_coder_inference/deploy.py

bash

./deploy/huggingface/qwen_coder_inference/connect_kitsune.sh "https://wiesio111-kitsune-qwen-coder.hf.space"

Pierwsze uruchomienie Space trwa kilka–kilkanaście minut (build llama.cpp + pobranie GGUF).