{ "schema_version": 1, "family": "controlfoley", "display_name": "ControlFoley", "description": "Multimodal Foley audio generation model.", "category": "audio_tools", "status": "experimental", "tasks": [ "sfx" ], "modes": [ "offline" ], "languages": [ "auto" ], "runtime": { "tags": [ "gguf" ] }, "capabilities": { "sfx": [ "prompt_generation" ] }, "ui": { "recommended_package": "controlfoley_large_44k_q8_0", "tags": [ "SFX", "GGUF" ], "docs": [ "docs/models/controlfoley.md", "docs/gguf.md" ] }, "sources": [ { "format": "gguf", "roots": { "model": ".", "weights": "$gguf" }, "files": { "open_clip_vocab": "model:open_clip_vocab.json", "open_clip_merges": "model:open_clip_merges.txt", "open_clip_tokenizer_config": "model:open_clip_tokenizer_config.json" }, "tensors": { "flow": { "source": "weights:", "prefix": "flow" }, "vae": { "source": "weights:", "prefix": "vae" }, "bigvgan": { "source": "weights:", "prefix": "bigvgan" }, "open_clip": { "source": "weights:", "prefix": "open_clip" }, "cav_mae": { "source": "weights:", "prefix": "cav_mae" }, "synchformer": { "source": "weights:", "prefix": "synchformer" }, "clap": { "source": "weights:", "prefix": "clap" }, "mert": { "source": "weights:", "prefix": "mert" }, "musicgen_style": { "source": "weights:", "prefix": "musicgen_style" } } } ], "options": { "request": [ { "name": "duration_sec", "type": "float", "required": false, "default": 8, "min": 0.64, "description": "Target temporal budget." }, { "name": "num_inference_steps", "type": "int", "required": false, "default": 25, "min": 1, "description": "Euler flow inference steps." }, { "name": "guidance_scale", "type": "float", "required": false, "default": 4.5, "min": 0, "description": "Classifier-free guidance strength. Must be > 0. Values strictly between 0 and 1 bypass CFG and negative-prompt conditioning." }, { "name": "negative_prompt", "type": "string", "required": false, "default": "", "description": "Negative text prompt used for the unconditional guidance text branch." }, { "name": "video_file", "type": "path", "required": false, "description": "Optional video path for video-conditioned Foley generation." }, { "name": "mask_away_clip", "type": "bool", "required": false, "default": false, "description": "Disable OpenCLIP video conditioning while keeping visual and sync conditioning." }, { "name": "seed", "type": "int", "required": false, "default": 42, "min": 0, "max": 4294967295, "description": "Generation seed." } ], "session": [ { "name": "weight_type", "type": "enum", "required": false, "preset": "weight_type_full", "default": "native", "description": "ControlFoley weight storage type." } ], "load": [] }, "package_defaults": { "download": { "kind": "huggingface_snapshot", "repo": "audio-cpp/audio.cpp-gguf", "revision": "main", "gated": false } }, "packages": [ { "id": "controlfoley_large_44k_f32", "display_name": "ControlFoley Large 44k F32 GGUF", "format": "gguf", "precision": "f32", "target_directory": "ControlFoley-GGUF", "files": [ "ControlFoley-GGUF/controlfoley-large-44k-f32.gguf" ], "strip_prefix": "ControlFoley-GGUF" }, { "id": "controlfoley_large_44k_q8_0", "display_name": "ControlFoley Large 44k Q8_0 GGUF", "default": true, "format": "gguf", "precision": "q8_0", "target_directory": "ControlFoley-GGUF", "files": [ "ControlFoley-GGUF/controlfoley-large-44k-q8_0.gguf" ], "strip_prefix": "ControlFoley-GGUF" } ], "dependencies": [] }