{"version":"1.0","type":"rich","title":"Deploying and Fine-Tuning an Uncensored DeepSeek R1 Distill Model on Google Cloud","description":"This example demonstrates a complete workflow for deploying an uncensored DeepSeek R1 Distill model to Google Cloud Run with GPU support and fine-tuning it using a multi-stage pipeline. It includes a FastAPI inference server with `/v1/inference` and `/v1/finetune` endpoints, Docker configuration, and a fine-tuning process using Hugging Face Transformers and TRL that handles cold start data, reasoning reinforcement learning, and final tuning phases. The code shows concrete deployment commands using `gcloud run deploy` with `nvidia-l4` GPU specs, and processes JSONL training files for model customization.","provider_name":"VizHub AI Studio","provider_url":"https://next.vizhub.com","html":"<iframe src=\"https://next.vizhub.com/viz/gist-9720a91f5b14c5a357d2c171ec65ce9c?mode=embed\" width=\"960\" height=\"500\" scrolling=\"no\" frameborder=\"1\"></iframe>","width":960,"height":500,"cache_age":3600}