Update Readme

Update Readme and mobile UI

Update Readme and mobile UI

Update Readme and field notes

Update Readme

Update README with new dual-backend info and add local GPU install script

Remove Nemotron from Space backend due to unresolvable Mamba dependencies

Fallback to slow Mamba path for Nemotron

Remove mamba deps to fix build

Add Mamba dependencies for Nemotron

Support Nemotron and Gemma base models in ZeroGPU backend

Re-enable ZeroGPU for chat and transcribe endpoints

Use merged model directly — no PEFT/LoRA needed at runtime

Fix generate crash: extract input_ids tensor from BatchEncoding

CPU inference: remove @spaces.GPU, fail-safe LoRA loading

Split model loading: base at startup, LoRA inside @spaces.GPU

Dual backend: transformers+ZeroGPU on Spaces, llama.cpp locally

Update Readme and GPU usage

Update Readme and GPU usage

Update Readme and GPU usage

Update Readme

Updates for GPU usage

Auto-download chat models on startup

Auto-download chat models on startup

Add GPU decorator

Fix OOM by using pre-compiled wheels for llama.cpp

Update README

Update README

Update README with Backyard AI track

Initial deploy to Hugging Face Space