cocktailpeanut

Self-hosted AI workspace for local-first chat, agents, tools, memory, research, documents, email, and model endpoint management.

Run image generation, GGUF language models, Whisper speech recognition, and Kokoro speech synthesis locally from one offline studio.

Prompt-first symbolic-music studio powered by live TypeSafe Jev decisions. Describe music in words, listen to an editable composition, keep shaping the same piece. Requires a TYPESAFE_API_KEY for live mode; a no-credit fixture mode is built in.

The most powerful and modular diffusion model GUI, api and backend with a graph/nodes interface. https://github.com/comfyanonymous/ComfyUI

Generate editable Ideogram JSON prompts from uploaded images.

Run and train AI models with a unified local interface. https://github.com/unslothai/unsloth

Launcher for Stable Audio 3 Small Music, Small SFX, and NVIDIA Medium using public cocktailpeanut Hugging Face mirrors. https://github.com/Stability-AI/stable-audio-3

Describe any song in plain English, compose a structured caption and lyrics locally, and generate it with MiniMax Music 3.

[NVIDIA Only] Dead simple web UI for training FLUX LoRA with LOW VRAM support (From 12GB)

Browser-based AI audio DAW for Stable Audio 3 with text-to-audio, inpainting, LoRA training, FFmpeg effects, waveform editing, sequencer, piano roll, and persistent library. https://github.com/gantasmo/stabledaw

Open source UI for ACE-Step 1.5 music generation.

1 Click Installer for Retrieval-based-Voice-Conversion-WebUI (https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI)

Browse ComfyUI workflow templates and manage missing model files in the folders exposed by a running local ComfyUI server.

Local multi-instrument audio-to-MIDI transcription from Kyutai and Mirelo.

Image-to-3D Gaussian splat generation from VAST-AI-Research. Requires a CUDA-capable GPU.

A Python framework for AI-driven character animation using neural networks.

Local-first voice synthesis studio powered by Qwen3-TTS.

One-click launcher for the original HiDream-O1-Image web UI using lazy-downloaded drbaph Dev or Full FP8 checkpoints through a root FP8 runner. Requires an NVIDIA CUDA GPU.

Text-to-Video (T2V) generation framework from Vchitect https://github.com/Vchitect/LaVie

Describe an image, get a 100% schema-valid Ideogram 4 JSON prompt — generated fully locally with an embedded llama.cpp (no Ollama or LM Studio required).