run training as detached subprocess to survive Gradio session timeout

use bfloat16 precision for training to halve RAM usage

stop ace-server during training to free RAM, restart after, add log visibility

pre-download training checkpoints at build time

pin diffusers==0.30.3 for torch 2.4.x compat

pin torchaudio==2.4.0 (before torchcodec default backend)

force uninstall torchcodec (torchaudio dependency, broken on Ubuntu 22.04)

monkey-patch torchaudio.load to use soundfile backend

remove torchcodec (needs FFmpeg 5.x), keep soundfile backend

add torchcodec + ffmpeg for torchaudio audio loading

add soundfile + libsndfile1 for torchaudio backend

add all training deps: diffusers lightning numpy tensorboard

add torchaudio dep for training preprocess

add loguru dep for training preprocess

install gradio[mcp] for mcp_server support

add git to runtime image for ace-step source clone

add LoRA training, fix css kwarg

full app: CLI + Gradio + training placeholder + MCP

rewrite app.py for ace-server HTTP API, no torch

copy all shared libs including versioned symlinks

add pkg-config for BLAS detection

switch to acestep.cpp GGUF: XL turbo Q4_K_M + 4B LM Q5_K_M

clickable title

slim title

compact grid UI, 4B default, show training model

force float32 on training model for CPU

fix: LoRA rank param is 'r' not 'rank'

fix training: use LoRAConfigV2 instead of Union type alias

add torchcodec for training audio loading

step 7: real app.py

step 6: +numba (all deps)

step 5: +vector-quantize-pytorch

step 4: +peft +torchvision

step 3: core deps without heavy ones

step label

step 2: torch cpu-only

step 1: bare gradio

initial commit