Fabio
c36510c21e
Some checks failed
Linux Build / Linux cpu (push) Has been cancelled
Linux Build / Linux cpu-clang (push) Has been cancelled
Linux Build / Linux vulkan (push) Has been cancelled
mac-build / macOS CPU (push) Has been cancelled
Nix Build / Nix (cpu) (push) Has been cancelled
Nix Build / Nix (vulkan) (push) Has been cancelled
windows-build / Windows CPU (push) Has been cancelled
324 lines
81 KiB
JSON
324 lines
81 KiB
JSON
{
|
||
"host": "127.0.0.1",
|
||
"port": 8088,
|
||
"device": 0,
|
||
"threads": 1,
|
||
"_comment": "Native WebUI catalog for model families enabled in registry.cpp. The embedded server loads one selected model or precision at a time and resolves paths relative to the configured models root. Missing packages appear as not installed; the Models page invokes the repository-level model_manager_v2.py through the server to install spec-backed GGUF packages in the background. 'task' must be one of: tts, asr, vad, diar, sep, gen, clon, vc, s2s, align, vdes, spk, svc. 'download_id' is a model_specs package id and is omitted for bundled assets such as silero_vad. Optional per-entry keys include input_hint, default_options, session_options, and min_vram_gb. min_vram_gb is an estimated minimum CUDA VRAM value used only for UI guidance.",
|
||
|
||
"models": [
|
||
{ "id": "ace-step", "display_name": "ACE-Step 1.5 (music gen)", "family": "ace_step", "path": "models/Ace-Step1.5", "task": "gen", "mode": "offline", "download_id": "ace_step", "default_text": "upbeat pop music with bright vocals and energetic drums", "session_options": { "ace_step.mem_saver": "true", "ace_step.dit_weight_type": "q8_0", "ace_step.text_encoder_weight_type": "q8_0", "ace_step.planner_weight_type": "q8_0" }, "min_vram_gb": 8 },
|
||
{ "id": "ming3-tts", "display_name": "Ming 3.0 TTS 0.5B", "family": "ming3_tts", "path": "models/Ming-3.0-TTS-0.5B-GGUF", "task": "tts", "mode": "offline", "download_id": "ming_3_0_tts_0_5b_bf16", "default_options": { "language": "en" }, "default_text": "Good evening. The harbor is quiet tonight." },
|
||
{ "id": "ming3-tts-clone", "display_name": "Ming 3.0 TTS 0.5B Clone", "family": "ming3_tts", "path": "models/Ming-3.0-TTS-0.5B-GGUF", "task": "clon", "mode": "offline", "download_id": "ming_3_0_tts_0_5b_bf16", "default_options": { "language": "en" }, "input_hint_en": "Upload a reference voice. Its transcript is optional and enables acoustic-prefix conditioning." },
|
||
{ "id": "ming3-tts-design", "display_name": "Ming 3.0 TTS 0.5B Voice Design", "family": "ming3_tts", "path": "models/Ming-3.0-TTS-0.5B-GGUF", "task": "vdes", "mode": "offline", "download_id": "ming_3_0_tts_0_5b_bf16", "default_options": { "language": "en" } },
|
||
{ "id": "ming3-audio-asr", "display_name": "Ming 3.0 Audio Tiny ASR", "family": "ming3_audio", "path": "models/Ming-3.0-Audio-Tiny-GGUF", "task": "asr", "mode": "offline", "download_id": "ming3_audio_tiny_bf16" },
|
||
{ "id": "ming3-audio-understanding", "display_name": "Ming 3.0 Audio Tiny Understanding", "family": "ming3_audio", "path": "models/Ming-3.0-Audio-Tiny-GGUF", "task": "asr", "workflow": "analysis", "mode": "offline", "download_id": "ming3_audio_tiny_bf16", "default_options": { "prompt": "Describe the sounds in this recording." } },
|
||
{ "id": "ming3-audio-response", "display_name": "Ming 3.0 Audio Tiny Spoken Response", "family": "ming3_audio", "path": "models/Ming-3.0-Audio-Tiny-GGUF", "task": "s2s", "mode": "offline", "download_id": "ming3_audio_tiny_bf16", "default_options": { "voice_id": "US_James", "language": "en" } },
|
||
{ "id": "ming3-audio-tts", "display_name": "Ming 3.0 Audio Tiny TTS", "family": "ming3_audio", "path": "models/Ming-3.0-Audio-Tiny-GGUF", "task": "tts", "mode": "offline", "download_id": "ming3_audio_tiny_bf16", "default_options": { "voice_id": "US_James", "language": "en" } },
|
||
{ "id": "ming3-audio-clone", "display_name": "Ming 3.0 Audio Tiny Clone", "family": "ming3_audio", "path": "models/Ming-3.0-Audio-Tiny-GGUF", "task": "clon", "mode": "offline", "download_id": "ming3_audio_tiny_bf16", "default_options": { "language": "en" }, "input_hint_en": "Upload a reference voice. Add its transcript to enable acoustic-prefix conditioning." },
|
||
{ "id": "xtts-v2", "display_name": "XTTS v2", "family": "xtts_v2", "path": "models/XTTS-v2-Q8", "task": "tts", "mode": "offline", "download_id": "xtts_v2_gguf_q8_0", "input_hint_en": "Upload a reference voice for multilingual speech synthesis." },
|
||
{ "id": "latentsync", "display_name": "LatentSync 1.6 Lip Sync", "family": "latentsync", "path": "models/LatentSync-1.6-GGUF", "task": "gen", "mode": "offline", "download_id": "latentsync_1_6_q8_0_f16", "arena": false, "input_hint_en": "Choose a face video and driving speech audio." },
|
||
{ "id": "apollo", "display_name": "Apollo", "family": "apollo", "path": "models/Apollo-GGUF", "task": "s2s", "mode": "offline", "download_id": "apollo_orig" },
|
||
{ "id": "ast-audioset", "display_name": "AST AudioSet", "family": "ast_audioset", "path": "models/AST-AudioSet-GGUF/ast-audioset-f32.gguf", "task": "cls", "mode": "offline", "download_id": "ast_audioset_f32" },
|
||
{ "id": "audio-flamingo-3", "display_name": "Audio Flamingo 3", "family": "audio_flamingo", "path": "models/Audio-Flamingo-GGUF/audio-flamingo-3-q4_k.gguf", "task": "asr", "mode": "offline", "download_id": "audio_flamingo_3_q4_k", "workflow": "analysis" },
|
||
{ "id": "audio-flamingo-next", "display_name": "Audio Flamingo Next", "family": "audio_flamingo", "path": "models/Audio-Flamingo-GGUF/audio-flamingo-next-q4_k.gguf", "task": "asr", "mode": "offline", "download_id": "audio_flamingo_next_q4_k", "workflow": "analysis" },
|
||
{ "id": "audio8-tts", "display_name": "Audio8 TTS Preview 0.6B (tts 克隆, GGUF Q8)", "display_name_en": "Audio8 TTS Preview 0.6B (tts + clone, GGUF Q8)", "family": "audio8_tts", "path": "models/Audio8-TTS-Preview-0.6B-GGUF", "task": "tts", "mode": "offline", "download_id": "audio8_tts_preview_0_6b_q8_0", "min_vram_gb": 4,
|
||
"input_hint": "**Audio8 TTS Preview 0.6B**：多语种 TTS / 零样本克隆（支持 yue/zh/nl/en/fr/de/it/ja/ko/pl/es/auto）；上传参考音色+参考文本即克隆，留空为普通 TTS；长文本自动分句。",
|
||
"input_hint_en": "**Audio8 TTS Preview 0.6B**: multilingual TTS and zero-shot clone (yue/zh/nl/en/fr/de/it/ja/ko/pl/es/auto). Upload a reference voice + transcript to clone; leave empty for plain TTS. Long text is chunked automatically." },
|
||
{ "id": "audiosr", "display_name": "AudioSR (audio super-resolution)", "family": "audiosr", "path": "models/AudioSR-GGUF/audiosr-basic-f32.gguf", "task": "s2s", "mode": "offline", "download_id": "audiosr_basic_f32", "min_vram_gb": 8,
|
||
"input_hint_en": "**AudioSR**: upload a source audio file to generate a super-resolved output." },
|
||
{ "id": "auk-edit", "display_name": "AuK (audio editing, community)", "family": "auk", "path": "models/AuK-Base-and-Flash-GGUF", "task": "gen", "mode": "offline", "arena": false, "download_id": "auk_base_f32", "default_text": "Remove the background noise and make the voice clearer.", "min_vram_gb": 20,
|
||
"input_hint_en": "**AuK audio editing** (community): upload source audio and describe the requested speech, singing, acoustic, paralinguistic, enhancement, or separation edit." },
|
||
{ "id": "auk-tts", "display_name": "AuK (speech generation, community)", "family": "auk", "path": "models/AuK-Base-and-Flash-GGUF", "task": "tts", "mode": "offline", "arena": false, "download_id": "auk_base_f32", "default_text": "The next train leaves in ten minutes.", "min_vram_gb": 20,
|
||
"input_hint_en": "**AuK** (community): enter text plus a voice description, or provide an optional reference voice for zero-shot speech generation. Select compatible Base or AuK-Flash, Qwen, and VAE components before loading." },
|
||
{ "id": "breeze-tts-clone", "display_name": "BreezeTTS 2 Clone", "family": "breeze_tts", "path": "models/Breeze-TTS-2-GGUF/breeze-tts-2-q8_0.gguf", "task": "clon", "mode": "offline", "download_id": "breeze_tts_2_q8_0", "min_vram_gb": 8,
|
||
"input_hint_en": "**BreezeTTS 2 Clone**: upload a reference voice and provide the matching reference transcript." },
|
||
{ "id": "breeze-tts", "display_name": "BreezeTTS 2 VoiceDesign", "family": "breeze_tts", "path": "models/Breeze-TTS-2-GGUF/breeze-tts-2-q8_0.gguf", "task": "vdes", "mode": "offline", "download_id": "breeze_tts_2_q8_0", "min_vram_gb": 8,
|
||
"input_hint_en": "**BreezeTTS 2 VoiceDesign**: enter text and describe the target voice in Model parameters. No reference voice is required." },
|
||
{ "id": "bs-roformer", "display_name": "BS-RoFormer (sep 人声分离)", "display_name_en": "BS-RoFormer (vocal separation)", "family": "bs_roformer", "path": "models/BS-RoFormer-ep368-GGUF/bs-roformer-ep368-q8_0.gguf", "task": "sep", "mode": "offline", "download_id": "bs_roformer_q8_0", "min_vram_gb": 3 },
|
||
{ "id": "canary-asr", "display_name": "Canary 180M Flash", "family": "canary_asr", "path": "models/Canary-180M-Flash-GGUF", "task": "asr", "mode": "offline", "download_id": "canary_180m_flash_f32" },
|
||
{ "id": "ced-base", "display_name": "CED Base", "family": "ced", "path": "models/CED-GGUF/ced-base-f32.gguf", "task": "cls", "mode": "offline", "download_id": "ced_base_f32" },
|
||
{ "id": "ced-mini", "display_name": "CED Mini", "family": "ced", "path": "models/CED-GGUF/ced-mini-f32.gguf", "task": "cls", "mode": "offline", "download_id": "ced_mini_f32" },
|
||
{ "id": "ced-small", "display_name": "CED Small", "family": "ced", "path": "models/CED-GGUF/ced-small-f32.gguf", "task": "cls", "mode": "offline", "download_id": "ced_small_f32" },
|
||
{ "id": "ced-tiny", "display_name": "CED Tiny", "family": "ced", "path": "models/CED-GGUF/ced-tiny-f32.gguf", "task": "cls", "mode": "offline", "download_id": "ced_tiny_f32" },
|
||
{ "id": "chatterbox-vc", "display_name": "Chatterbox (vc 声音转换)", "display_name_en": "Chatterbox (voice conversion)", "family": "chatterbox", "path": "models/chatterbox", "task": "vc", "mode": "offline", "download_id": "chatterbox", "min_vram_gb": 12,
|
||
"input_hint": "**Chatterbox VC**：上传源语音和目标音色参考；模型保留源语音内容，将说话人音色转换为目标音色，输出 24kHz 单声道。",
|
||
"input_hint_en": "**Chatterbox VC**: upload source speech and a target-voice reference. It preserves the source content and converts the speaker identity; output is 24 kHz mono." },
|
||
{ "id": "chatterbox", "display_name": "Chatterbox (voice clone)", "family": "chatterbox", "path": "models/chatterbox", "task": "clon", "mode": "offline", "download_id": "chatterbox", "min_vram_gb": 12 },
|
||
{ "id": "chatterbox-turbo", "display_name": "Chatterbox Turbo (tts)", "family": "chatterbox_turbo", "path": "models/Chatterbox-Turbo-GGUF/chatterbox-turbo-q8_0.gguf", "task": "tts", "mode": "offline", "download_id": "chatterbox_turbo_q8_0", "min_vram_gb": 4,
|
||
"input_hint_en": "**Chatterbox Turbo**: fast English TTS with the built-in voice. No reference voice is required." },
|
||
{ "id": "citrinet-asr", "display_name": "Citrinet ASR (asr)", "family": "citrinet_asr", "path": "models/citrinet", "task": "asr", "mode": "offline", "download_id": "citrinet_asr", "min_vram_gb": 2 },
|
||
{ "id": "cohere-asr", "display_name": "Cohere Transcribe", "family": "cohere_asr", "path": "models/Cohere-Transcribe-GGUF", "task": "asr", "mode": "offline", "download_id": "cohere_transcribe_bf16" },
|
||
{ "id": "r2t2-asr", "display_name": "Confucius4-R2T2 (asr, 实时流式)", "display_name_en": "Confucius4-R2T2 (asr, real-time streaming)", "family": "confucius4_r2t2", "path": "models/Confucius4-R2T2-GGUF/r2t2-q8_0.gguf", "task": "asr", "mode": "offline", "download_id": "confucius4_r2t2_q8_0", "min_vram_gb": 5,
|
||
"input_hint": "**Confucius4-R2T2**：网易有道实时流式 ASR，Qwen3-ASR-1.7B 微调，LSP 稳定前缀解码；提交文本永不回改，支持 80ms-2s 分块；Q8_0 GGUF（2.3G，自包含单文件），也支持 F16。", "input_hint_en": "**Confucius4-R2T2**: NetEase Youdao real-time streaming ASR, a Qwen3-ASR 1.7B fine-tune with Longest Stable Prefix decoding. Committed text is never revised; 80 ms-2 s chunks; Q8_0 GGUF (2.3 GB, self-contained single file), F16 also available." },
|
||
{ "id": "confucius4-tts", "display_name": "Confucius4-TTS (voice clone, GGUF)", "display_name_en": "Confucius4-TTS (voice clone, GGUF)", "family": "confucius4_tts", "path": "models/Confucius4-TTS-GGUF", "task": "clon", "mode": "offline", "download_id": "confucius4_tts_orig", "min_vram_gb": 8,
|
||
"input_hint": "**Confucius4-TTS**：需要参考音色；当前中文/英语路径更可靠，非中英语种仍在验证中。",
|
||
"input_hint_en": "**Confucius4-TTS**: requires a reference voice. Chinese/English are the most reliable paths; other languages are still being validated." },
|
||
{ "id": "controlfoley", "display_name": "ControlFoley (Foley/SFX)", "family": "controlfoley", "path": "models/ControlFoley-GGUF/controlfoley-large-44k-q8_0.gguf", "task": "gen", "mode": "offline", "download_id": "controlfoley_large_44k_q8_0", "min_vram_gb": 12,
|
||
"request_options": ["video"],
|
||
"input_hint_en": "**ControlFoley**: Foley generation from text, video, text+video, audio+video, or video only. Upload video in the Video field; upload source audio for AC-V2A." },
|
||
{ "id": "cosyvoice3", "display_name": "CosyVoice3 Clone", "family": "cosyvoice3", "path": "models/CosyVoice3-GGUF/cosyvoice3-q8_0.gguf", "task": "clon", "mode": "offline", "download_id": "cosyvoice3_q8_0", "min_vram_gb": 8,
|
||
"input_hint_en": "**CosyVoice3 Clone**: upload a reference voice and provide the matching reference transcript. Use `template_name` for zero-shot or cross-lingual requests." },
|
||
{ "id": "cosyvoice3-instruct", "display_name": "CosyVoice3 Instruct", "family": "cosyvoice3", "path": "models/CosyVoice3-GGUF/cosyvoice3-q8_0.gguf", "task": "tts", "mode": "offline", "download_id": "cosyvoice3_q8_0", "min_vram_gb": 8,
|
||
"input_hint_en": "**CosyVoice3 Instruct**: upload a reference voice, provide its transcript, and set `template_name=instruct` with an instruction." },
|
||
{ "id": "crisperwhisper", "display_name": "CrisperWhisper 2.0 Large", "family": "crisperwhisper", "path": "models/CrisperWhisper2.0-GGUF/crisperwhisper-2-large-q8_0.gguf", "task": "asr", "mode": "offline", "download_id": "crisperwhisper_2_large_q8_0" },
|
||
{ "id": "crisperwhisper-align", "display_name": "CrisperWhisper 2.0 Large (alignment)", "family": "crisperwhisper", "path": "models/CrisperWhisper2.0-GGUF/crisperwhisper-2-large-q8_0.gguf", "task": "align", "mode": "offline", "download_id": "crisperwhisper_2_large_q8_0", "workflow": "analysis" },
|
||
{ "id": "deepfilternet2", "display_name": "DeepFilterNet2", "family": "builtin_audio_utils", "task": "s2s", "workflow": "enhancement", "mode": "offline", "path": "models/Audio-Utilities/deepfilternet2/deepfilternet2.safetensors", "load_options": { "utility": "deepfilternet2" }, "weights_url": "https://raw.githubusercontent.com/0xShug0/audio.cpp/main/assets/framework/audio_utilities/deepfilternet2/deepfilternet2.safetensors" },
|
||
{ "id": "dots-tts-meanflow", "display_name": "DotTTS MeanFlow (tts + clone)", "family": "dots_tts", "path": "models/DotTTS-MF-GGUF", "task": "tts", "mode": "offline", "download_id": "dots_tts_mf_q8_0", "min_vram_gb": 8 },
|
||
{ "id": "dots-tts-soar", "display_name": "DotTTS SOAR (tts + clone)", "family": "dots_tts", "path": "models/DotTTS-SOAR-GGUF", "task": "tts", "mode": "offline", "download_id": "dots_tts_soar_q8_0", "min_vram_gb": 8 },
|
||
{ "id": "dramabox", "display_name": "DramaBox (tts 克隆, GGUF Q8)", "display_name_en": "DramaBox (tts + clone, GGUF Q8)", "family": "dramabox", "path": "models/DramaBox-GGUF", "task": "tts", "mode": "offline", "download_id": "dramabox_q8_0", "min_vram_gb": 16,
|
||
"input_hint": "**DramaBox**：英语 TTS / voice clone；上传参考音色可克隆，长文本建议写清楚说话人描述。",
|
||
"input_hint_en": "**DramaBox**: English TTS and voice clone. Upload a reference voice to clone; for long text, keep speaker wording explicit." },
|
||
{ "id": "echo-tts", "display_name": "Echo-TTS (voice clone)", "family": "echo_tts", "path": "models/Echo-TTS-GGUF", "task": "clon", "mode": "offline", "download_id": "echo_tts_q8_0", "min_vram_gb": 8,
|
||
"input_hint_en": "**Echo-TTS**: English zero-shot cloning at 44.1 kHz. Upload a reference voice -- no transcript needed. Output is CC-BY-NC-SA and may not be used commercially." },
|
||
{ "id": "firered-stream-vad", "display_name": "FireRed Stream-VAD", "family": "firered_vad", "path": "models/FireRedVAD-GGUF/firered-stream-vad-f32.gguf", "task": "vad", "mode": "offline", "download_id": "firered_stream_vad_f32" },
|
||
{ "id": "firered-vad", "display_name": "FireRed VAD", "family": "firered_vad", "path": "models/FireRedVAD-GGUF/firered-vad-f32.gguf", "task": "vad", "mode": "offline", "download_id": "firered_vad_f32" },
|
||
{ "id": "firered-audio-asr", "display_name": "FireRedAudio (ASR / audio QA)", "family": "firered_audio", "path": "models/FireRedAudio-GGUF/firered-audio-q8_0.gguf", "task": "asr", "mode": "offline", "download_id": "firered_audio_q8_0", "min_vram_gb": 10,
|
||
"input_hint_en": "**FireRedAudio ASR**: upload audio, then use the text box as the transcription or audio-understanding instruction." },
|
||
{ "id": "firered-audio-acoustic-edit", "display_name": "FireRedAudio Acoustic Edit", "family": "firered_audio", "path": "models/FireRedAudio-GGUF/firered-audio-q8_0.gguf", "task": "gen", "mode": "offline", "download_id": "firered_audio_q8_0", "min_vram_gb": 10,
|
||
"input_hint_en": "**FireRedAudio Acoustic Edit**: upload source audio and use a trained acoustic instruction such as `shift the pitch by 3 steps`." },
|
||
{ "id": "firered-audio-tts", "display_name": "FireRedAudio Clone", "family": "firered_audio", "path": "models/FireRedAudio-GGUF/firered-audio-q8_0.gguf", "task": "clon", "mode": "offline", "download_id": "firered_audio_q8_0", "min_vram_gb": 10,
|
||
"input_hint_en": "**FireRedAudio Clone**: upload a reference voice and provide the matching reference transcript. For no-reference voice design, use the VoiceDesign entry." },
|
||
{ "id": "firered-audio-semantic-edit", "display_name": "FireRedAudio Semantic Edit", "family": "firered_audio", "path": "models/FireRedAudio-GGUF/firered-audio-q8_0.gguf", "task": "gen", "mode": "offline", "download_id": "firered_audio_q8_0", "min_vram_gb": 10,
|
||
"input_hint_en": "**FireRedAudio Semantic Edit**: upload source audio and describe the content edit in Model parameters." },
|
||
{ "id": "firered-audio-vdesign", "display_name": "FireRedAudio VoiceDesign", "family": "firered_audio", "path": "models/FireRedAudio-GGUF/firered-audio-q8_0.gguf", "task": "vdes", "mode": "offline", "download_id": "firered_audio_q8_0", "min_vram_gb": 10,
|
||
"input_hint_en": "**FireRedAudio VoiceDesign**: describe the target voice in Voice description, then enter the text to synthesize." },
|
||
{ "id": "fireredtts3-base", "display_name": "FireRedTTS3 Base (voice clone)", "family": "fireredtts3", "path": "models/FireRedTTS3-Base-GGUF/fireredtts3-base-q8_0.gguf", "task": "clon", "mode": "offline", "download_id": "fireredtts3_base_q8_0", "min_vram_gb": 8,
|
||
"input_hint_en": "**FireRedTTS3 Base**: zero-shot voice cloning. Upload a reference voice and provide the matching reference transcript." },
|
||
{ "id": "fireredtts3-instruct", "display_name": "FireRedTTS3 Instruct Clone", "family": "fireredtts3", "path": "models/FireRedTTS3-Instruct-GGUF/fireredtts3-instruct-q8_0.gguf", "task": "clon", "mode": "offline", "download_id": "fireredtts3_instruct_q8_0", "min_vram_gb": 8,
|
||
"input_hint_en": "**FireRedTTS3 Instruct Clone**: official Instruct `generate_tts` path. Upload a reference voice and provide the matching reference transcript. For no-reference voice design, use the VoiceDesign entry." },
|
||
{ "id": "fireredtts3-instruct-vdesign", "display_name": "FireRedTTS3 Instruct VoiceDesign", "family": "fireredtts3", "path": "models/FireRedTTS3-Instruct-GGUF/fireredtts3-instruct-q8_0.gguf", "task": "vdes", "mode": "offline", "download_id": "fireredtts3_instruct_q8_0", "min_vram_gb": 8,
|
||
"input_hint_en": "**FireRedTTS3 VoiceDesign**: describe the target voice in Voice description, then enter the text to synthesize." },
|
||
{"id":"fish-drama3","display_name":"Fish Audio Drama 3","family":"fish_drama3","path":"models/Drama-3-GGUF/drama-3-backbone-4bit-experts-Q4_K_S.gguf","task":"tts","mode":"offline","arena":false,"download_id":"fish_drama3_q4_backbone","min_vram_gb":8,"session_options":{"fish_drama3.disk_experts":"true","fish_drama3.pin_experts":"true","fish_drama3.expert_cache_size":"8","fish_drama3.cpu_expert_layers":"0","fish_drama3.codec_backend":"same","fish_drama3.context_length":"4096"},"request_options":["reference_text","multi_reference_cond","text_chunk_size","temperature","top_p","top_k","max_tokens","seed"]},
|
||
{"id":"fish-drama3-small","display_name":"Fish Audio Drama3-small","family":"fish_drama3","path":"models/Drama3-small-GGUF/drama3-small-IQ3_XXS.gguf","task":"tts","mode":"offline","arena":false,"download_id":"fish_drama3_small_iq3_xxs","min_vram_gb":7,"session_options":{"fish_drama3.disk_experts":"true","fish_drama3.pin_experts":"true","fish_drama3.expert_cache_size":"8","fish_drama3.cpu_expert_layers":"0","fish_drama3.codec_backend":"same","fish_drama3.context_length":"4096"}},
|
||
{ "id": "fish-audio-s2-pro", "display_name": "Fish Audio S2 Pro (tts 克隆/控制标记, GGUF Q8)", "display_name_en": "Fish Audio S2 Pro (tts + clone/control tags, GGUF Q8)", "family": "fish_audio", "path": "models/Fish-Audio-S2-Pro-GGUF", "task": "tts", "mode": "offline", "download_id": "fish_audio_s2_pro", "min_vram_gb": 8,
|
||
"input_hint": "**Fish Audio S2 Pro**：Q8_0 GGUF 包；中英+自动语种；上传参考音色即克隆；正文里可写行内控制标记（如 (laugh)）。",
|
||
"input_hint_en": "**Fish Audio S2 Pro**: Q8_0 GGUF package; English/Chinese plus auto language. Upload a reference voice to clone; inline control tags such as (laugh) can be written in the text." },
|
||
{ "id": "flashsr", "display_name": "FlashSR", "family": "builtin_audio_utils", "task": "s2s", "workflow": "enhancement", "mode": "offline", "path": "models/Audio-Utilities/flashsr/flashsr.safetensors", "load_options": { "utility": "flashsr" }, "weights_url": "https://raw.githubusercontent.com/0xShug0/audio.cpp/main/assets/framework/audio_utilities/flashsr/flashsr.safetensors" },
|
||
{ "id": "fun-asr-nano", "display_name": "Fun-ASR-Nano 2512 (asr, GGUF Q8)", "display_name_en": "Fun-ASR-Nano 2512 (asr, GGUF Q8)", "family": "fun_asr_nano", "path": "models/Fun-ASR-Nano-2512-GGUF", "task": "asr", "mode": "offline", "download_id": "fun_asr_nano_2512_q8_0", "min_vram_gb": 4,
|
||
"input_hint": "**Fun-ASR-Nano**：轻量离线 ASR；支持 auto/中文/英语/日语。",
|
||
"input_hint_en": "**Fun-ASR-Nano**: lightweight offline ASR; supports auto, Chinese, English and Japanese." },
|
||
{ "id": "gigaam-multilingual", "display_name": "GigaAM Multilingual", "family": "gigaam_asr", "task": "asr", "mode": "offline", "path": "models/GigaAM-ASR-GGUF/gigaam-multilingual-ctc-f16.gguf", "download_id": "gigaam_multilingual_ctc_f16" },
|
||
{ "id": "gigaam-multilingual-large", "display_name": "GigaAM Multilingual Large", "family": "gigaam_asr", "task": "asr", "mode": "offline", "path": "models/GigaAM-ASR-GGUF/gigaam-multilingual-large-ctc-f16.gguf", "download_id": "gigaam_multilingual_large_ctc_f16" },
|
||
{ "id": "gigaam-v3-ctc", "display_name": "GigaAM v3 CTC", "family": "gigaam_asr", "task": "asr", "mode": "offline", "path": "models/GigaAM-ASR-GGUF/gigaam-v3-ctc-f16.gguf", "download_id": "gigaam_v3_ctc_f16" },
|
||
{ "id": "gigaam-v3-e2e-ctc", "display_name": "GigaAM v3 E2E CTC", "family": "gigaam_asr", "task": "asr", "mode": "offline", "path": "models/GigaAM-ASR-GGUF/gigaam-v3-e2e-ctc-f16.gguf", "download_id": "gigaam_v3_e2e_ctc_f16" },
|
||
{ "id": "gigaam-v3-e2e-rnnt", "display_name": "GigaAM v3 E2E RNN-T", "family": "gigaam_asr", "task": "asr", "mode": "offline", "path": "models/GigaAM-ASR-GGUF/gigaam-v3-e2e-rnnt-f16.gguf", "download_id": "gigaam_v3_e2e_rnnt_f16" },
|
||
{ "id": "gigaam-v3-rnnt", "display_name": "GigaAM v3 RNN-T", "family": "gigaam_asr", "task": "asr", "mode": "offline", "path": "models/GigaAM-ASR-GGUF/gigaam-v3-rnnt-f16.gguf", "download_id": "gigaam_v3_rnnt_f16" },
|
||
{ "id": "glm-tts", "display_name": "GLM-TTS (tts 克隆, 社区)", "display_name_en": "GLM-TTS (tts + clone, community)", "family": "glm_tts", "path": "models/GLM-TTS", "task": "tts", "mode": "offline", "download_id": "glm_tts", "min_vram_gb": 8,
|
||
"input_hint": "**GLM-TTS**（社区模型）：中英 TTS / voice clone；上传参考音色即克隆。",
|
||
"input_hint_en": "**GLM-TTS** (community): Chinese/English TTS and voice clone. Upload a reference voice to clone." },
|
||
{ "id": "gradium-phonon-31m-en", "display_name": "Gradium Phonon 31m English (tts)", "family": "gradium_phonon", "path": "models/phonon-31m-en", "task": "tts", "mode": "offline", "download_id": "gradium_phonon_31m_en_q8_0", "min_vram_gb": 1 },
|
||
{ "id": "granite5asr", "display_name": "Granite Speech 5.0 470M TurboCTC (asr)", "display_name_en": "Granite Speech 5.0 470M TurboCTC (asr)", "family": "granite5asr", "path": "models/Granite-Speech-5.0-470M-TurboCTC-GGUF/granite-speech-5.0-470m-turboctc-q8_0.gguf", "task": "asr", "mode": "offline", "download_id": "granite5asr_q8_0", "min_vram_gb": 4,
|
||
"input_hint": "**Granite Speech 5.0 TurboCTC**：IBM 470M 英语 ASR；超快 Conformer CTC 转写；支持长音频自动分段与流式模式。",
|
||
"input_hint_en": "**Granite Speech 5.0 TurboCTC**: IBM 470M English ASR with ultra-fast Conformer CTC architecture, supporting long-form audio segmentation and streaming mode." },
|
||
{ "id": "gtcrn-dns3", "display_name": "GTCRN DNS3", "family": "builtin_audio_utils", "task": "s2s", "workflow": "enhancement", "mode": "offline", "path": "models/Audio-Utilities/gtcrn/gtcrn_dns3.safetensors", "load_options": { "utility": "gtcrn_dns3" }, "weights_url": "https://raw.githubusercontent.com/0xShug0/audio.cpp/main/assets/framework/audio_utilities/gtcrn/gtcrn_dns3.safetensors" },
|
||
{ "id": "gtcrn-streaming", "display_name": "GTCRN Streaming", "family": "builtin_audio_utils", "task": "s2s", "workflow": "enhancement", "mode": "offline", "path": "models/Audio-Utilities/gtcrn/gtcrn_streaming.safetensors", "load_options": { "utility": "gtcrn_streaming" }, "weights_url": "https://raw.githubusercontent.com/0xShug0/audio.cpp/main/assets/framework/audio_utilities/gtcrn/gtcrn_streaming.safetensors" },
|
||
{ "id": "gtcrn-vctk", "display_name": "GTCRN VCTK", "family": "builtin_audio_utils", "task": "s2s", "workflow": "enhancement", "mode": "offline", "path": "models/Audio-Utilities/gtcrn/gtcrn_vctk.safetensors", "load_options": { "utility": "gtcrn_vctk" }, "weights_url": "https://raw.githubusercontent.com/0xShug0/audio.cpp/main/assets/framework/audio_utilities/gtcrn/gtcrn_vctk.safetensors" },
|
||
{ "id": "heartmula", "display_name": "HeartMuLa 3B (music gen)", "family": "heartmula", "path": "models/HeartMuLa", "task": "gen", "mode": "offline", "download_id": "heartmula", "session_options": { "heartmula.mem_saver": "true" }, "min_vram_gb": 24 },
|
||
{ "id": "higgs-audio-stt", "display_name": "Higgs Audio v3 STT (asr, 英语)", "display_name_en": "Higgs Audio v3 STT (asr, English)", "family": "higgs_audio_stt", "path": "models/higgs-audio-v3-stt", "task": "asr", "mode": "offline", "download_id": "higgs_audio_stt", "min_vram_gb": 8,
|
||
"input_hint": "**Higgs Audio STT**：英语转写；可在文本框填指令（默认相当于 Transcribe the speech.）；离线模式自动切分长音频。", "input_hint_en": "**Higgs Audio STT**: English transcription. The text box accepts an instruction; offline mode automatically chunks long audio." },
|
||
{ "id": "higgs-audio-tts", "display_name": "Higgs Audio v3 TTS 4B (tts 克隆, GGUF Q8)", "display_name_en": "Higgs Audio v3 TTS 4B (tts + clone, GGUF Q8)", "family": "higgs_audio_tts", "path": "models/Higgs-Audio-v3-TTS-4B-GGUF", "task": "tts", "mode": "offline", "download_id": "higgs_audio_v3_tts_4b", "min_vram_gb": 6,
|
||
"input_hint": "**Higgs Audio v3 TTS**：Q8_0 GGUF 包（权重已量化，不用再设 weight_type）；上传参考音色即声音克隆，留空用默认音色；长文本自动分段。",
|
||
"input_hint_en": "**Higgs Audio v3 TTS**: Q8_0 GGUF package (already quantized — no weight_type needed). Upload a reference voice to clone, or leave it empty for the default voice; long text is chunked automatically." },
|
||
{ "id": "htdemucs", "display_name": "HTDemucs (sep 音源分离)", "display_name_en": "HTDemucs (source separation)", "family": "htdemucs", "path": "models/htdemucs", "task": "sep", "mode": "offline", "download_id": "htdemucs", "min_vram_gb": 3 },
|
||
{ "id": "htdemucs_6stems", "display_name": "HTDemucs 6-stem (sep 音源分离)", "display_name_en": "HTDemucs 6-stem (source separation)", "family": "htdemucs_6stems", "path": "models/htdemucs_6stems", "task": "sep", "mode": "offline", "download_id": "htdemucs_6stems", "min_vram_gb": 4 },
|
||
{ "id": "hviske-asr", "display_name": "Hviske v5.3 (asr, 丹麦语)", "display_name_en": "Hviske v5.3 (asr, Danish)", "family": "hviske_asr", "path": "models/hviske-v5.3", "task": "asr", "mode": "offline", "download_id": "hviske_asr", "min_vram_gb": 6,
|
||
"input_hint": "**Hviske ASR**：丹麦语专用；模型侧自动分段。", "input_hint_en": "**Hviske ASR**: dedicated Danish transcription with automatic model-side segmentation." },
|
||
{ "id": "hviske-v6", "display_name": "Hviske v6 (Danish ASR)", "family": "hviske_asr_v6", "path": "models/Hviske-v6-GGUF/hviske-v6-bf16.gguf", "task": "asr", "mode": "offline", "download_id": "hviske_asr_v6_bf16" },
|
||
{ "id": "index-echo-2b", "display_name": "Index-Echo S2TT 2B", "family": "index_echo", "path": "models/Index-Echo-S2TT-GGUF/index-echo-s2tt-2b-orig.gguf", "task": "asr", "mode": "offline", "download_id": "index_echo_s2tt_2b_orig" },
|
||
{ "id": "index-echo-9b", "display_name": "Index-Echo S2TT 9B", "family": "index_echo", "path": "models/Index-Echo-S2TT-GGUF/index-echo-s2tt-9b-orig.gguf", "task": "asr", "mode": "offline", "download_id": "index_echo_s2tt_9b_orig" },
|
||
{ "id": "index-tts2", "display_name": "IndexTTS2 (tts 中英克隆+情感)", "display_name_en": "IndexTTS2 (tts, zh/en clone + emotion)", "family": "index_tts2", "path": "models/IndexTTS-2", "task": "tts", "mode": "offline", "download_id": "index_tts2", "min_vram_gb": 8 },
|
||
{ "id": "index-tts2.5", "display_name": "IndexTTS2.5 (tts 多语种克隆+情感, GGUF Q8)", "display_name_en": "IndexTTS2.5 (tts, zh/en/ja/es/ar clone + emotion, GGUF Q8)", "family": "index_tts2", "path": "models/IndexTTS2.5-GGUF", "task": "tts", "mode": "offline", "download_id": "index_tts2_5_q8_0", "min_vram_gb": 8,
|
||
"input_hint": "**IndexTTS2.5**：中/英/日/西/阿零样本克隆；上传参考音色即克隆；可在『其它参数(JSON)』里传 `lang`（默认 auto：含汉字按中文，否则按英文）与情感选项。许可证为 bilibili Model Use License（非 OSI），商用前请确认条款。",
|
||
"input_hint_en": "**IndexTTS2.5**: zero-shot cloning in zh/en/ja/es/ar. Upload a reference voice to clone; pass `lang` (default auto: zh when the text contains Han characters, otherwise en) and emotion options through the JSON box. Weights are under the bilibili Model Use License (not OSI-approved) — check terms before commercial use." },
|
||
{ "id": "inflect-v2", "display_name": "Inflect Micro v2 (tts 英语, 社区)", "display_name_en": "Inflect Micro v2 (en tts, community)", "family": "inflect_v2", "path": "models/Inflect-Micro-v2", "task": "tts", "mode": "offline", "download_id": "inflect_micro_v2", "min_vram_gb": 2,
|
||
"input_hint": "**Inflect Micro v2**（社区模型）：英语离线 TTS；Micro 是默认包，Nano 可通过模型管理器另装后手动选择路径。",
|
||
"input_hint_en": "**Inflect Micro v2** (community): English offline TTS. Micro is the default package; Nano can be installed separately and selected manually." },
|
||
{ "id": "irodori-tts-v3-500m", "display_name": "Irodori-TTS 500M v3 (tts 日语)", "display_name_en": "Irodori-TTS 500M v3 (ja tts)", "family": "irodori_tts", "path": "models/Irodori-TTS-500M-v3-GGUF", "task": "tts", "mode": "offline", "download_id": "irodori_tts_500m_v3_q8_0", "min_vram_gb": 4 },
|
||
{ "id": "irodori-tts-v3-vdesign", "display_name": "Irodori-TTS 600M v3 VoiceDesign (vdes 日语)", "display_name_en": "Irodori-TTS 600M v3 VoiceDesign (ja vdes)", "family": "irodori_tts", "path": "models/Irodori-TTS-600M-v3-VoiceDesign-GGUF", "task": "vdes", "mode": "offline", "download_id": "irodori_tts_600m_v3_voicedesign_q8_0", "min_vram_gb": 4,
|
||
"input_hint": "**Irodori-TTS v3 VoiceDesign**（日语）：『音色描述』用日语 caption 描述音色（如「落ち着いた大人の男性。深く響く声。」），文本填要念的日语内容，无需参考音频。", "input_hint_en": "**Irodori-TTS v3 VoiceDesign**: provide a Japanese voice caption and Japanese synthesis text. No reference recording is required." },
|
||
{ "id": "irodori-tts", "display_name": "Irodori-TTS v4.1 Small (tts 日语, GGUF Q8)", "display_name_en": "Irodori-TTS v4.1 Small (ja tts, GGUF Q8)", "family": "irodori_tts", "path": "models/Irodori-TTS-v4-Small-GGUF", "task": "tts", "mode": "offline", "download_id": "irodori_tts_v4_small_q8_0", "min_vram_gb": 4,
|
||
"input_hint": "**Irodori-TTS v4.1 Small**：日语 TTS；可不上传参考音色直接生成，也可上传参考音色进行克隆；可在声音设计页用日语 caption 描述音色。",
|
||
"input_hint_en": "**Irodori-TTS v4.1 Small**: Japanese TTS. Generate without a reference voice, clone from an uploaded reference, or use the voice-design page with a Japanese voice caption." },
|
||
{ "id": "irodori-tts-vdesign", "display_name": "Irodori-TTS v4.1 Small VoiceDesign (vdes 日语, GGUF Q8)", "display_name_en": "Irodori-TTS v4.1 Small VoiceDesign (ja vdes, GGUF Q8)", "family": "irodori_tts", "path": "models/Irodori-TTS-v4-Small-GGUF", "task": "vdes", "mode": "offline", "download_id": "irodori_tts_v4_small_q8_0", "min_vram_gb": 4,
|
||
"input_hint": "**Irodori-TTS v4.1 VoiceDesign**（日语）：『音色描述』用日语 caption 描述音色（如「落ち着いた大人の男性。深く響く声。」），文本填要念的日语内容，无需参考音频。", "input_hint_en": "**Irodori-TTS v4.1 VoiceDesign**: provide a Japanese voice caption and Japanese synthesis text. No reference recording is required." },
|
||
{ "id": "kitten-tts2", "display_name": "Kitten TTS 2", "family": "kitten_tts2", "path": "models/kitten-tts2/kitten-tts2-q8-v2-multilingual.gguf", "task": "tts", "mode": "offline", "download_id": "kitten_tts2_q8_v2", "min_vram_gb": 7,
|
||
"input_hint": "Multilingual speech: choose a matching language preset or supply a reference clip and its transcript. Write numbers and abbreviations as spoken words." },
|
||
{ "id": "kitten-tts2-20261021", "display_name": "Kitten TTS 2 (2026-10-21)", "family": "kitten_tts2", "path": "models/KittenTTS2-GGUF", "task": "tts", "mode": "offline", "download_id": "kitten_tts2_20261021_q8_0", "input_hint_en": "Updated multilingual checkpoint. Choose Q8 or Q4, then a voice preset or reference recording." },
|
||
{ "id": "kitten-tts", "display_name": "KittenTTS Mini 0.8 (tts 英语, 社区)", "display_name_en": "KittenTTS Mini 0.8 (en tts, community)", "family": "kitten_tts", "path": "models/KittenTTS-GGUF/kitten-tts-mini-0.8-orig.gguf", "task": "tts", "mode": "offline", "download_id": "kitten_tts_mini_0_8_orig", "min_vram_gb": 1,
|
||
"input_hint": "**KittenTTS Mini 0.8**（社区模型）：轻量英语 TTS，内置八种音色，不需要上传参考音频。",
|
||
"input_hint_en": "**KittenTTS Mini 0.8** (community): lightweight English TTS with eight built-in voices; no reference audio is needed." },
|
||
{ "id": "kokoro-tts", "display_name": "Kokoro 82M (tts, preset voices)", "family": "kokoro_tts", "path": "models/Kokoro-82M-GGUF/kokoro-82m-q8_0.gguf", "task": "tts", "mode": "offline", "download_id": "kokoro_82m_q8_0", "min_vram_gb": 1,
|
||
"input_hint_en": "**Kokoro 82M**: multilingual TTS with built-in preset voices. Choose a language and voice; no reference audio is needed." },
|
||
{ "id": "kroko-asr", "display_name": "Kroko Community ASR (asr, GGUF Q8)", "display_name_en": "Kroko Community ASR (asr, GGUF Q8)", "family": "kroko_asr", "path": "models/Kroko-ASR-GGUF", "task": "asr", "mode": "offline", "download_id": "kroko_asr_community_q8_0", "min_vram_gb": 4,
|
||
"input_hint": "**Kroko Community ASR**：GGUF Q8 包；离线转写，支持时间戳。",
|
||
"input_hint_en": "**Kroko Community ASR**: GGUF Q8 package for offline transcription with timestamps." },
|
||
{ "id": "kugelaudio", "display_name": "KugelAudio-0-Open", "family": "kugelaudio", "path": "models/KugelAudio-0-Open-GGUF/kugelaudio-0-open-q8_0.gguf", "task": "tts", "mode": "offline", "download_id": "kugelaudio_0_open_q8_0", "builtin_voices": ["default", "clear", "english_female", "english_male"], "default_voice": "default" },
|
||
{ "id": "lfm2-audio-asr", "display_name": "LFM2.5-Audio 1.5B (asr, community)", "family": "lfm2_audio", "path": "models/LFM2.5-Audio-1.5B-GGUF", "task": "asr", "mode": "offline", "download_id": "lfm2_audio_1_5b_q8_0", "min_vram_gb": 4,
|
||
"input_hint_en": "**LFM2.5-Audio** (community): Liquid AI's English speech recognition from Liquid's published GGUFs. Audio longer than 30 s is split at pauses. Leave language empty." },
|
||
{ "id": "lfm2-audio-s2s", "display_name": "LFM2.5-Audio 1.5B (s2s, community)", "family": "lfm2_audio", "path": "models/LFM2.5-Audio-1.5B-GGUF", "task": "s2s", "mode": "offline", "download_id": "lfm2_audio_1_5b_q8_0", "min_vram_gb": 4, "builtin_voices": [], "default_voice": "",
|
||
"input_hint_en": "**LFM2.5-Audio** (community): Liquid AI's English spoken chat from Liquid's published GGUFs. Record or choose a spoken question under Source audio and press Run: the reply streams in as speech and text, and each Run continues the conversation until New conversation. Replies use liquid-audio's system prompt, `Respond with interleaved text and audio.`" },
|
||
{ "id": "lfm2-audio-tts", "display_name": "LFM2.5-Audio 1.5B (tts, community)", "family": "lfm2_audio", "path": "models/LFM2.5-Audio-1.5B-GGUF", "task": "tts", "mode": "offline", "download_id": "lfm2_audio_1_5b_q8_0", "min_vram_gb": 4, "default_text": "The next train leaves in ten minutes.",
|
||
"input_hint_en": "**LFM2.5-Audio** (community): Liquid AI's English speech from Liquid's published GGUFs, with four built-in voices (US or UK, male or female); no reference audio." },
|
||
{ "id": "lfm2-audio-jp-asr", "display_name": "LFM2.5-Audio 1.5B JP (asr, community)", "family": "lfm2_audio", "path": "models/LFM2.5-Audio-1.5B-JP-GGUF", "task": "asr", "mode": "offline", "download_id": "lfm2_audio_1_5b_jp_q8_0", "min_vram_gb": 4,
|
||
"input_hint_en": "**LFM2.5-Audio JP** (community): Liquid AI's Japanese speech recognition from Liquid's published GGUFs. Audio longer than 30 s is split at pauses. Leave language empty." },
|
||
{ "id": "lfm2-audio-jp-s2s", "display_name": "LFM2.5-Audio 1.5B JP (s2s, community)", "family": "lfm2_audio", "path": "models/LFM2.5-Audio-1.5B-JP-GGUF", "task": "s2s", "mode": "offline", "download_id": "lfm2_audio_1_5b_jp_q8_0", "min_vram_gb": 4, "builtin_voices": [], "default_voice": "",
|
||
"input_hint_en": "**LFM2.5-Audio JP** (community): Liquid AI's Japanese spoken chat from Liquid's published GGUFs. Record or choose a spoken question under Source audio and press Run: the reply streams in as speech and text, and each Run continues the conversation until New conversation. Replies use liquid-audio's system prompt, `Respond with interleaved text and audio.`" },
|
||
{ "id": "lfm2-audio-jp-tts", "display_name": "LFM2.5-Audio 1.5B JP (tts, community)", "family": "lfm2_audio", "path": "models/LFM2.5-Audio-1.5B-JP-GGUF", "task": "tts", "mode": "offline", "download_id": "lfm2_audio_1_5b_jp_q8_0", "min_vram_gb": 4, "builtin_voices": [], "default_voice": "", "default_text": "今日は天気がいいので、公園まで散歩に行きました。",
|
||
"input_hint_en": "**LFM2.5-Audio JP** (community): Liquid AI's Japanese speech from Liquid's published GGUFs, in the checkpoint's one voice; no reference audio." },
|
||
{ "id": "lfm3-audio-asr", "display_name": "LFM3-Audio 2B (asr, community)", "family": "lfm3_audio", "path": "models/LFM3-Audio-2B-GGUF/LFM3-Audio-2B-Q8_0.gguf", "task": "asr", "mode": "offline", "download_id": "lfm3_audio_2b_q8_0", "min_vram_gb": 4,
|
||
"input_hint_en": "**LFM3-Audio** (community): Liquid AI's LFM3-Audio-2B from Liquid's published GGUFs. Speech recognition in 18 languages: leave language empty or set auto to detect it. The chat parameter answers the spoken input as text instead of transcribing it." },
|
||
{ "id": "lfm3-audio-s2s", "display_name": "LFM3-Audio 2B (s2s, community)", "family": "lfm3_audio", "path": "models/LFM3-Audio-2B-GGUF/LFM3-Audio-2B-Q8_0.gguf", "task": "s2s", "mode": "offline", "download_id": "lfm3_audio_2b_q8_0", "min_vram_gb": 4, "builtin_voices": [], "default_voice": "",
|
||
"input_hint_en": "**LFM3-Audio** (community): Liquid AI's LFM3-Audio-2B from Liquid's published GGUFs. Record or choose a spoken question under Source audio and press Run: the reply streams in as speech and text, and each Run continues the conversation until New conversation. Replies use the system prompt `Respond with interleaved text and audio.`" },
|
||
{ "id": "lfm3-audio-tts", "display_name": "LFM3-Audio 2B (tts, community)", "family": "lfm3_audio", "path": "models/LFM3-Audio-2B-GGUF/LFM3-Audio-2B-Q8_0.gguf", "task": "tts", "mode": "offline", "download_id": "lfm3_audio_2b_q8_0", "min_vram_gb": 4, "default_text": "The next train leaves in ten minutes.",
|
||
"input_hint_en": "**LFM3-Audio** (community): Liquid AI's LFM3-Audio-2B from Liquid's published GGUFs. Speech in 38 built-in voices across 19 languages; no reference audio." },
|
||
{ "id": "soulx-flashhead", "display_name": "SoulX-FlashHead Lite (community)", "family": "soulx_flashhead", "path": "models/SoulX-FlashHead-GGUF", "task": "gen", "mode": "offline", "download_id": "soulx_flashhead_lite_q8_0", "min_vram_gb": 10,
|
||
"input_hint_en": "Audio-driven portrait video. Voice chat in Apps combines this model with LFM3-Audio and an ASR model. cuDNN is recommended for CUDA builds." },
|
||
{ "id": "liveavatar", "display_name": "LiveAvatar (audio to video, 社区)", "display_name_en": "LiveAvatar (audio to video, community)", "family": "liveavatar", "path": "models/LiveAvatar-GGUF", "task": "gen", "mode": "offline", "download_id": "liveavatar_nvfp4_lora", "default_text": "A cheerful green cyclops baker in a warm rustic bakery, smiling and speaking to the camera while holding a tray of freshly baked pies. High-quality animated film style, expressive face, natural gestures, warm cinematic lighting.", "min_vram_gb": 24,
|
||
"input_hint": "**LiveAvatar**（社区模型）：上传驱动音频和参考人物图像，再输入场景描述以生成动画视频。",
|
||
"input_hint_en": "**LiveAvatar** (community): upload driving audio and a reference portrait, then describe the subject and scene to generate an animated video." },
|
||
{ "id": "outetts", "display_name": "Llama-OuteTTS 1.0 1B (tts 克隆, 社区)", "display_name_en": "Llama-OuteTTS 1.0 1B (tts + clone, community)", "family": "outetts", "path": "models/Llama-OuteTTS-1.0-1B", "task": "tts", "mode": "offline", "download_id": "outetts_1_0_1b", "min_vram_gb": 4,
|
||
"input_hint": "**OuteTTS 1.0 1B**（社区模型）：23 种语言，DAC 编解码；上传参考音色即克隆。",
|
||
"input_hint_en": "**OuteTTS 1.0 1B** (community): 23 languages, IBM DAC codec. Upload a reference voice to clone." },
|
||
{ "id": "magenta-rt2", "display_name": "Magenta RealTime 2 (music gen)", "family": "magenta_realtime_2", "path": "models/Magenta-RealTime-2-GGUF", "task": "gen", "mode": "offline", "download_id": "magenta_realtime_2_small_f16", "default_text": "disco funk" },
|
||
{ "id": "magpie-tts", "display_name": "MagpieTTS Multilingual 357M v2607 (tts, preset voices)", "display_name_en": "MagpieTTS Multilingual 357M v2607 (tts, preset voices)", "family": "magpie_tts", "path": "models/MagpieTTS-Multilingual-357M-GGUF", "task": "tts", "mode": "offline", "download_id": "magpie_tts_q8_0", "min_vram_gb": 4,
|
||
"input_hint": "**MagpieTTS**：多语种离线 TTS；使用打包 speaker map 里的 `voice_id`，不需要上传参考音频。当前 GGUF 包不包含日语 phoneme 表，因此日语路径不可用。",
|
||
"input_hint_en": "**MagpieTTS**: multilingual offline TTS with packaged speaker prompts selected by `voice_id`; no reference upload is needed. The current GGUF package does not include the Japanese phoneme table, so Japanese is not available." },
|
||
{ "id": "marblenet-vad", "display_name": "MarbleNet VAD (vad, bundled)", "family": "marblenet_vad", "path": "assets/framework/models/marblenet_vad", "task": "vad", "mode": "offline", "min_vram_gb": 1 },
|
||
{ "id": "maya1", "display_name": "Maya1", "family": "maya1", "task": "tts", "mode": "offline", "path": "models/Maya1-GGUF/maya1-q8_0.gguf", "download_id": "maya1_q8_0", "arena": false },
|
||
{ "id": "meanvc2", "display_name": "MeanVC2 (voice conversion)", "display_name_en": "MeanVC2 (voice conversion)", "family": "meanvc2", "path": "models/MeanVC2-GGUF/meanvc2-120ms-40ms-fp32.gguf", "task": "vc", "mode": "offline", "download_id": "meanvc2_120ms_40ms_f32", "min_vram_gb": 6,
|
||
"input_hint": "**MeanVC2**：上传源语音和目标音色参考；默认 120 ms / 40 ms checkpoint 使用 F32 GGUF。",
|
||
"input_hint_en": "**MeanVC2**: upload source speech and a target-voice reference. The default 120 ms / 40 ms checkpoint uses F32 GGUF." },
|
||
{ "id": "mel-band-roformer", "display_name": "Mel-Band RoFormer (sep 人声分离)", "display_name_en": "Mel-Band RoFormer (vocal separation)", "family": "mel_band_roformer", "path": "models/mel-roformer-mlx", "task": "sep", "mode": "offline", "download_id": "mel_band_roformer", "min_vram_gb": 3 },
|
||
{ "id": "micro-wake-word", "display_name": "microWakeWord (Okay Nabu)", "family": "micro_wake_word", "path": "models/MicroWakeWord-GGUF/micro-wake-word-okay-nabu-f32.gguf", "task": "wake", "mode": "offline", "download_id": "micro_wake_word_okay_nabu_f32" },
|
||
{ "id": "midashenglm-gen", "display_name": "MiDashengLM-Gen (audio generation)", "family": "midashenglm_gen", "path": "models/MiDashengLM-Gen-GGUF/midashenglm-gen-q8_0.gguf", "task": "gen", "mode": "offline", "download_id": "midashenglm_gen_q8_0", "min_vram_gb": 8,
|
||
"input_hint_en": "**MiDashengLM-Gen**: text-conditioned audio generation. The duration field controls the generation budget." },
|
||
{ "id": "minimax-h3", "display_name": "MiniMax-H3 Q4 (sound generation)", "family": "minimax_h3", "path": "models/MiniMax-H3-Q4-GGUF/dit.gguf", "task": "gen", "mode": "offline", "download_id": "minimax_h3", "min_vram_gb": 20,
|
||
"default_options": { "num_inference_steps": 12, "height": 32, "width": 32, "num_frames": 241, "guidance_scale": 1.0, "dit_acceleration": "none", "return_video": false },
|
||
"input_hint_en": "MiniMax-H3 uses a joint audio/video DiT. The default Q4 DiT is the quality-first choice; the optional CUDA-only INT8 ConvRot DiT trades slightly more VRAM for higher speed. Native Studio uses 12 denoising steps, a 32x32 latent canvas, quality-first full-DiT execution, and disables video decoding for practical audio-only generation on a 24 GB GPU." },
|
||
{ "id": "minimax-music3", "display_name": "MiniMax-Music3 (song gen)", "family": "minimax_music3", "path": "models/MiniMax-Music3-GGUF", "task": "gen", "mode": "offline", "download_id": "minimax_music3_q4_0", "min_vram_gb": 12 },
|
||
{ "id": "miocodec", "display_name": "MioCodec (vc; codec dependency)", "family": "miocodec", "path": "models/MioCodec-25Hz-44.1kHz-v2", "task": "vc", "mode": "offline", "download_id": "miocodec_25hz_44k_v2", "min_vram_gb": 3 },
|
||
{ "id": "miotts", "display_name": "MioTTS 1.7B (tts; needs MioCodec)", "family": "miotts", "path": "models/MioTTS-1.7B", "task": "tts", "mode": "offline", "download_id": "miotts_1_7b", "min_vram_gb": 8 },
|
||
{ "id": "moonshine-asr-medium", "display_name": "Moonshine Streaming Medium (asr, GGUF Q8)", "display_name_en": "Moonshine Streaming Medium (asr, GGUF Q8)", "family": "moonshine_asr", "path": "models/Moonshine-Streaming-GGUF/moonshine-streaming-medium-q8_0.gguf", "task": "asr", "mode": "offline", "download_id": "moonshine_streaming_medium_q8_0", "min_vram_gb": 2,
|
||
"input_hint": "**Moonshine Streaming Medium**：英语 ASR，GGUF Q8 包；支持离线与流式模式。",
|
||
"input_hint_en": "**Moonshine Streaming Medium**: English ASR GGUF Q8 package with offline and streaming support." },
|
||
{ "id": "moonshine-asr-small", "display_name": "Moonshine Streaming Small (asr, GGUF Q8)", "display_name_en": "Moonshine Streaming Small (asr, GGUF Q8)", "family": "moonshine_asr", "path": "models/Moonshine-Streaming-GGUF/moonshine-streaming-small-q8_0.gguf", "task": "asr", "mode": "offline", "download_id": "moonshine_streaming_small_q8_0", "min_vram_gb": 2,
|
||
"input_hint": "**Moonshine Streaming Small**：英语 ASR，GGUF Q8 包；支持离线与流式模式。",
|
||
"input_hint_en": "**Moonshine Streaming Small**: English ASR GGUF Q8 package with offline and streaming support." },
|
||
{ "id": "moonshine-asr-tiny", "display_name": "Moonshine Streaming Tiny (asr, GGUF Q8)", "display_name_en": "Moonshine Streaming Tiny (asr, GGUF Q8)", "family": "moonshine_asr", "path": "models/Moonshine-Streaming-GGUF/moonshine-streaming-tiny-q8_0.gguf", "task": "asr", "mode": "offline", "download_id": "moonshine_streaming_tiny_q8_0", "min_vram_gb": 1,
|
||
"input_hint": "**Moonshine Streaming Tiny**：英语 ASR，轻量 GGUF Q8 包；支持离线与流式模式。",
|
||
"input_hint_en": "**Moonshine Streaming Tiny**: lightweight English ASR GGUF Q8 package with offline and streaming support." },
|
||
{ "id": "moss-transcribe-diarize", "display_name": "MOSS-Transcribe-Diarize", "family": "moss_transcribe_diarize", "path": "models/MOSS-Transcribe-Diarize-GGUF", "task": "asr", "mode": "offline", "download_id": "moss_transcribe_diarize_bf16" },
|
||
{ "id": "moss-tts-v15", "display_name": "MOSS-TTS v1.5 8B (tts 克隆, 社区)", "display_name_en": "MOSS-TTS v1.5 8B (tts + clone, community)", "family": "moss_tts_v15", "path": "models/MOSS-TTS-v1.5-GGUF/moss_tts_v15_q8_0_codec_f16.gguf", "task": "tts", "mode": "offline", "download_id": "moss_tts_v15_q8_0_codec_f16", "min_vram_gb": 18,
|
||
"input_hint": "**MOSS-TTS v1.5 8B**（社区模型）：中英 TTS；上传短参考音频进行声音克隆，留空则使用模型生成的音色。Q8 克隆约需 18 GB 显存。",
|
||
"input_hint_en": "**MOSS-TTS v1.5 8B** (community): English/Chinese TTS. Upload a short reference clip for voice cloning, or leave it empty for a model-generated voice. Q8 cloning needs about 18 GB VRAM." },
|
||
{ "id": "moss-tts-local", "display_name": "MOSS-TTS-Local v1.5 (tts)", "family": "moss_tts_local", "path": "models/MOSS-TTS-Local-Transformer-v1.5", "task": "tts", "mode": "offline", "download_id": "moss_tts_local_v1_5", "min_vram_gb": 8 },
|
||
{ "id": "moss-tts-nano", "display_name": "MOSS-TTS-Nano 100M (tts)", "family": "moss_tts_nano", "path": "models/MOSS-TTS-Nano-100M", "task": "tts", "mode": "offline", "download_id": "moss_tts_nano_100m", "min_vram_gb": 2 },
|
||
{ "id": "moss-ttsd", "display_name": "MOSS-TTSD", "family": "moss_ttsd", "task": "tts", "mode": "offline", "path": "models/MOSS-TTSD-GGUF/moss_ttsd_q8_0_codec_f16.gguf", "download_id": "moss_ttsd_q8_0_codec_f16", "arena": false, "default_text": "[S1] Hello, how are you? [S2] I am doing well, thank you." },
|
||
{ "id": "moss-voicegen", "display_name": "MOSS-VoiceGenerator", "family": "moss_voicegen", "task": "vdes", "mode": "offline", "path": "models/MOSS-VoiceGenerator-GGUF/moss_voicegen_bf16_codec_f16_decode.gguf", "download_id": "moss_voicegen_bf16_codec_f16_decode", "arena": false },
|
||
{ "id": "mossformer2", "display_name": "MossFormer2 (two-speaker separation)", "family": "mossformer2", "path": "models/MossFormer2-GGUF/mossformer2-ss-16k-f32.gguf", "task": "sep", "mode": "offline", "download_id": "mossformer2_ss_16k_f32" },
|
||
{ "id": "muscriptor-small", "display_name": "MuScriptor Small (audio to MIDI)", "family": "muscriptor", "path": "models/MuScriptor-Small-GGUF", "task": "midi", "mode": "offline", "download_id": "muscriptor_small_f32", "min_vram_gb": 4 },
|
||
{ "id": "nemotron-3-diar", "display_name": "Nemotron 3 Diarization", "family": "nemotron_3_diar", "path": "models/Nemotron-3-Diarization-GGUF/nemotron-3-diarization-bf16.gguf", "task": "diar", "mode": "offline", "download_id": "nemotron_3_diar_bf16", "min_vram_gb": 2,
|
||
"input_hint_en": "**Nemotron 3 Diarization**: upload speech audio to produce speaker turns for up to eight speakers. It supports offline and streaming diarization." },
|
||
{ "id": "nemotron-asr", "display_name": "Nemotron 3.5 ASR 0.6B (asr, 100+语种)", "display_name_en": "Nemotron 3.5 ASR 0.6B (asr, 100+ languages)", "family": "nemotron_asr", "path": "models/nemotron-3.5-asr-streaming-0.6b", "task": "asr", "mode": "offline", "download_id": "nemotron_asr", "min_vram_gb": 4,
|
||
"input_hint": "**Nemotron ASR**：100+ 语种，语种码为 BCP-47（如 en-US / zh-CN），留空=auto；模型自带长音频处理。", "input_hint_en": "**Nemotron ASR**: supports more than 100 languages using BCP-47 codes such as en-US or zh-CN. Leave language blank for automatic detection; long audio is handled by the model." },
|
||
{ "id": "neutts-2e", "display_name": "NeuTTS 2E (tts, preset voices)", "family": "neutts", "path": "models/NeuTTS-2E-GGUF", "task": "tts", "mode": "offline", "download_id": "neutts_2e_orig", "min_vram_gb": 4 },
|
||
{ "id": "niagara-asr-19m", "display_name": "Niagara ASR 19M (asr)", "display_name_en": "Niagara ASR 19M (asr)", "family": "niagara_asr", "path": "models/Niagara-ASR-GGUF/niagara-19m-batch.en-f32.gguf", "task": "asr", "mode": "offline", "download_id": "niagara_19m_f32", "min_vram_gb": 1,
|
||
"input_hint": "**Niagara ASR 19M**：ABR 英语离线 ASR，F32 GGUF 权重。", "input_hint_en": "**Niagara ASR 19M**: ABR English offline ASR with F32 GGUF weights." },
|
||
{ "id": "niagara-asr-38m", "display_name": "Niagara ASR 38M (asr)", "display_name_en": "Niagara ASR 38M (asr)", "family": "niagara_asr", "path": "models/Niagara-ASR-GGUF/niagara-38m-batch.en-f32.gguf", "task": "asr", "mode": "offline", "download_id": "niagara_38m_f32", "min_vram_gb": 1,
|
||
"input_hint": "**Niagara ASR 38M**：ABR 英语离线 ASR，F32 GGUF 权重。", "input_hint_en": "**Niagara ASR 38M**: ABR English offline ASR with F32 GGUF weights." },
|
||
{ "id": "omnivoice", "display_name": "OmniVoice (tts)", "family": "omnivoice", "path": "models/OmniVoice", "task": "tts", "mode": "offline", "download_id": "omnivoice", "min_vram_gb": 10 },
|
||
{ "id": "open-wake-word", "display_name": "OpenWakeWord (Alexa)", "family": "open_wake_word", "path": "models/OpenWakeWord-GGUF/open-wake-word-alexa-f32.gguf", "task": "wake", "mode": "offline", "download_id": "open_wake_word_alexa_f32" },
|
||
{ "id": "orukeet", "display_name": "Orukeet r3 (asr, Parakeet 微调)", "display_name_en": "Orukeet r3 (asr, Parakeet fine-tune)", "family": "parakeet_tdt", "path": "models/Orukeet-GGUF/orukeet-q8_0.gguf", "task": "asr", "mode": "offline", "download_id": "orukeet_q8_0", "min_vram_gb": 4,
|
||
"input_hint": "**Orukeet r3**：Parakeet-TDT 0.6B v3 微调权重，同引擎/同分词器；CC BY-SA 4.0，留空=自动语种。",
|
||
"input_hint_en": "**Orukeet r3**: fine-tuned Parakeet-TDT 0.6B v3 weights on the same engine and tokenizer; CC BY-SA 4.0, leave language empty for auto." },
|
||
{ "id": "owsm-base", "display_name": "OWSM v4 Base 102M", "family": "owsm", "path": "models/OWSM-GGUF/owsm-v4-base-102m-f32.gguf", "task": "asr", "mode": "offline", "download_id": "owsm_v4_base_102m_f32" },
|
||
{ "id": "owsm-medium", "display_name": "OWSM v4 Medium 1B", "family": "owsm", "path": "models/OWSM-GGUF/owsm-v4-medium-1b-f32.gguf", "task": "asr", "mode": "offline", "download_id": "owsm_v4_medium_1b_f32" },
|
||
{ "id": "owsm-small", "display_name": "OWSM v4 Small 370M", "family": "owsm", "path": "models/OWSM-GGUF/owsm-v4-small-370m-f32.gguf", "task": "asr", "mode": "offline", "download_id": "owsm_v4_small_370m_f32" },
|
||
{ "id": "owsm-ctc", "display_name": "OWSM-CTC v4 1B", "family": "owsm_ctc", "path": "models/OWSM-CTC-GGUF/owsm-ctc-v4-1b-f32.gguf", "task": "asr", "mode": "offline", "download_id": "owsm_ctc_v4_1b_f32" },
|
||
{ "id": "parakeet-tdt", "display_name": "Parakeet-TDT 0.6B v3 (asr, 流式)", "display_name_en": "Parakeet-TDT 0.6B v3 (asr + streaming)", "family": "parakeet_tdt", "path": "models/parakeet-tdt-0.6b-v3", "task": "asr", "mode": "offline", "download_id": "parakeet_tdt", "min_vram_gb": 4,
|
||
"input_hint": "**Parakeet-TDT**：离线/长音频/流式 ASR；支持多种欧洲语言，留空=自动。",
|
||
"input_hint_en": "**Parakeet-TDT**: offline, long-form and streaming ASR for many European languages; leave language empty for auto." },
|
||
{ "id": "personaplex", "display_name": "PersonaPlex 7B v1 (speech conversation)", "display_name_en": "PersonaPlex 7B v1 (speech conversation)", "family": "personaplex", "path": "models/PersonaPlex-GGUF", "task": "s2s", "mode": "offline", "download_id": "personaplex_7b_v1_q4_k", "min_vram_gb": 8,
|
||
"request_options": ["voice_id", "system_prompt", "temperature", "text_temperature", "top_k", "text_top_k", "do_sample", "seed"],
|
||
"input_hint": "**PersonaPlex**：上传用户语音，模型返回语音回复；文本框可填写 assistant system/persona prompt；`voice_id` 选择打包音色，也可上传参考音色覆盖。",
|
||
"input_hint_en": "**PersonaPlex**: upload user speech and receive a spoken response. The text box provides the assistant system/persona prompt; `voice_id` selects a packaged voice, and an uploaded reference voice overrides it." },
|
||
{ "id": "piper-tts", "display_name": "Piper Lessac Medium (tts 英语, 社区)", "display_name_en": "Piper Lessac Medium (en tts, community)", "family": "piper_tts", "path": "models/Piper-TTS-GGUF/piper-en-us-lessac-medium-orig.gguf", "task": "tts", "mode": "offline", "download_id": "piper_lessac_medium_orig", "min_vram_gb": 1,
|
||
"input_hint": "**Piper Lessac Medium**（社区模型）：轻量英语 TTS，使用打包的 Lessac 音色，不需要上传参考音频。",
|
||
"input_hint_en": "**Piper Lessac Medium** (community): lightweight English TTS with the packaged Lessac voice; no reference audio is needed." },
|
||
{ "id": "plapre", "display_name": "Plapre 3 (Danish TTS)", "family": "plapre", "path": "models/Plapre-3-GGUF/plapre-3-bf16.gguf", "task": "tts", "mode": "offline", "download_id": "plapre_3_bf16", "default_text": "Godmorgen. Solen skinner, og vi tager en tur til havnen." },
|
||
{ "id": "pocket-tts", "display_name": "Pocket TTS (tts)", "family": "pocket_tts", "path": "models/pocket-tts", "task": "tts", "mode": "offline", "download_id": "pocket_tts", "min_vram_gb": 2 },
|
||
{ "id": "pulsevad-2.1k", "display_name": "PulseVAD 2.1K", "family": "pulsevad", "path": "models/PulseVAD-GGUF", "task": "vad", "mode": "offline", "download_id": "pulsevad_2_1k_f32" },
|
||
{ "id": "pulsevad-81k", "display_name": "PulseVAD 81K", "family": "pulsevad", "path": "models/PulseVAD-GGUF", "task": "vad", "mode": "offline", "download_id": "pulsevad_81k_f32" },
|
||
{ "id": "qwen3-forced-aligner", "display_name": "Qwen3 Forced Aligner (align)", "family": "qwen3_forced_aligner", "path": "models/Qwen3-ForcedAligner-0.6B", "task": "align", "mode": "offline", "download_id": "qwen3_forced_aligner_0_6b", "min_vram_gb": 3 },
|
||
{ "id": "qwen3-asr", "display_name": "Qwen3-ASR 0.6B (asr)", "family": "qwen3_asr", "path": "models/Qwen3-ASR-0.6B", "task": "asr", "mode": "offline", "download_id": "qwen3_asr_0_6b", "min_vram_gb": 3 },
|
||
{ "id": "qwen3-asr-1.7b", "display_name": "Qwen3-ASR 1.7B HF (asr)", "family": "qwen3_asr", "path": "models/Qwen3-ASR-1.7B-hf", "task": "asr", "mode": "offline", "download_id": "qwen3_asr_1_7b_hf", "min_vram_gb": 6,
|
||
"input_hint": "**Qwen3-ASR 1.7B**（HF 原生权重，免转换）：精度高于 0.6B；长音频自动分段转写；8G 卡显存偏紧，长音频建议先短段试跑。", "input_hint_en": "**Qwen3-ASR 1.7B**: native Hugging Face weights with no conversion required. It is more accurate than the 0.6B model and automatically chunks long audio; test short clips first on an 8 GB GPU." },
|
||
{ "id": "qwen3-tts", "display_name": "Qwen3-TTS 0.6B (tts)", "family": "qwen3_tts", "path": "models/Qwen3-TTS-12Hz-0.6B-Base", "task": "tts", "mode": "offline", "download_id": "qwen3_tts_0_6b_base", "min_vram_gb": 5 },
|
||
{ "id": "qwen3-tts-1.7b", "display_name": "Qwen3-TTS 1.7B Base (tts)", "family": "qwen3_tts", "path": "models/Qwen3-TTS-12Hz-1.7B-Base", "task": "tts", "mode": "offline", "download_id": "qwen3_tts_1_7b_base", "min_vram_gb": 8 },
|
||
{ "id": "qwen3-tts-1.7b-custom", "display_name": "Qwen3-TTS 1.7B CustomVoice (tts)", "family": "qwen3_tts", "path": "models/Qwen3-TTS-12Hz-1.7B-CustomVoice", "task": "tts", "mode": "offline", "download_id": "qwen3_tts_1_7b_custom_voice", "min_vram_gb": 8 },
|
||
{ "id": "qwen3-tts-1.7b-vdesign", "display_name": "Qwen3-TTS 1.7B VoiceDesign (vdes)", "family": "qwen3_tts", "path": "models/Qwen3-TTS-12Hz-1.7B-VoiceDesign", "task": "vdes", "mode": "offline", "download_id": "qwen3_tts_1_7b_voice_design", "min_vram_gb": 8,
|
||
"input_hint": "**Qwen3-TTS VoiceDesign**：在『音色描述』里用文字描述想要的声音（如“低沉磁性的中年男声，语速偏慢”），配上要念的文本即可，无需参考音频。", "input_hint_en": "**Qwen3-TTS VoiceDesign**: describe the desired voice, then enter the text to synthesize. No reference recording is required." },
|
||
{ "id": "reuse", "display_name": "RE-USE", "family": "reuse", "path": "models/RE-USE-GGUF/reuse-f32.gguf", "task": "s2s", "mode": "offline", "download_id": "reuse_f32", "workflow": "enhancement" },
|
||
{ "id": "rnnoise", "display_name": "RNNoise", "family": "builtin_audio_utils", "task": "s2s", "workflow": "enhancement", "mode": "offline", "path": "models/Audio-Utilities/rnnoise/rnnoise10Gb_15.safetensors", "load_options": { "utility": "rnnoise" }, "weights_url": "https://raw.githubusercontent.com/0xShug0/audio.cpp/main/assets/framework/audio_utilities/rnnoise/rnnoise10Gb_15.safetensors" },
|
||
{ "id": "rvc", "display_name": "RVC (vc, GGUF F16)", "display_name_en": "RVC (voice conversion, GGUF F16)", "family": "rvc", "path": "models/RVC-GGUF", "task": "vc", "mode": "offline", "download_id": "rvc_f16", "min_vram_gb": 4,
|
||
"input_hint": "**RVC**：所选 GGUF 即目标音色；上传源语音即可转换；索引/音高等选项可用 JSON 传。",
|
||
"input_hint_en": "**RVC**: the selected GGUF is the target voice; upload source speech to convert. Index and pitch options can be passed through the JSON box." },
|
||
{ "id": "sam-audio-base", "display_name": "SAM Audio Base", "family": "sam_audio", "task": "s2s", "workflow": "separation", "mode": "offline", "path": "models/SAM-Audio-GGUF/sam-audio-base-q8_0.gguf", "download_id": "sam_audio_base_q8_0", "default_text": "man speaking" },
|
||
{ "id": "sam-audio-large", "display_name": "SAM Audio Large", "family": "sam_audio", "task": "s2s", "workflow": "separation", "mode": "offline", "path": "models/SAM-Audio-GGUF/sam-audio-large-q8_0.gguf", "download_id": "sam_audio_large_q8_0", "default_text": "man speaking" },
|
||
{ "id": "sam-audio-small", "display_name": "SAM Audio Small", "family": "sam_audio", "task": "s2s", "workflow": "separation", "mode": "offline", "path": "models/SAM-Audio-GGUF/sam-audio-small-q8_0.gguf", "download_id": "sam_audio_small_q8_0", "default_text": "man speaking" },
|
||
{ "id": "samsone-99m", "display_name": "SAMSONE 99M", "family": "samsone", "task": "asr", "workflow": "analysis", "arena": false, "mode": "offline", "path": "models/Samsone-GGUF/samsone-99m-q8_0.gguf", "download_id": "samsone_99m_q8_0" },
|
||
{ "id": "samsone-134m", "display_name": "SAMSONE 134M", "family": "samsone", "task": "asr", "workflow": "analysis", "arena": false, "mode": "offline", "path": "models/Samsone-GGUF/samsone-134m-q8_0.gguf", "download_id": "samsone_134m_q8_0" },
|
||
{ "id": "samsone-356m", "display_name": "SAMSONE 356M", "family": "samsone", "task": "asr", "workflow": "analysis", "arena": false, "mode": "offline", "path": "models/Samsone-GGUF/samsone-356m-q8_0.gguf", "download_id": "samsone_356m_q8_0" },
|
||
{ "id": "sanotts", "display_name": "sanoTTS voice family (tts, community)", "family": "sanotts", "path": "models/sanoTTS-heart-nano-GGUF/heart-nano-f32.gguf", "task": "tts", "mode": "offline", "download_id": "sanotts_heart_nano_orig", "min_vram_gb": 1 },
|
||
{ "id": "seed-vc-svc", "display_name": "Seed-VC (svc 歌声转换)", "display_name_en": "Seed-VC (singing voice conversion)", "family": "seed_vc", "path": "models/SeedVC-MLX", "task": "svc", "mode": "offline", "download_id": "seed_vc", "min_vram_gb": 4,
|
||
"input_hint": "**Seed-VC 歌声转换 (svc)**：上传源歌声 + 目标歌手参考音色，默认 route=v1_svc（带 F0 条件）。可在『其它参数(JSON)』里调 `auto_f0_adjust` / `semi_tone_shift` / `f0_condition`。", "input_hint_en": "**Seed-VC singing conversion**: upload source singing and a target-singer reference. The default route is v1_svc with F0 conditioning." },
|
||
{ "id": "seed-vc", "display_name": "Seed-VC (vc 语音转换)", "display_name_en": "Seed-VC (voice conversion)", "family": "seed_vc", "path": "models/SeedVC-MLX", "task": "vc", "mode": "offline", "download_id": "seed_vc", "min_vram_gb": 4 },
|
||
{ "id": "sense-asr", "display_name": "SenseVoice-Small (asr, 流式, 社区)", "display_name_en": "SenseVoice-Small (asr + streaming, community)", "family": "sense_asr", "path": "models/SenseVoice-Small-GGUF", "task": "asr", "mode": "offline", "download_id": "sensevoice_small_q8", "min_vram_gb": 4,
|
||
"input_hint": "**SenseVoice-Small**（社区模型）：多语种 ASR，事件/情感/语言标签，ITN 可开关；离线与流式模式。",
|
||
"input_hint_en": "**SenseVoice-Small** (community): multilingual ASR with event/emotion/language tags, optional ITN; offline and streaming modes." },
|
||
{ "id": "sheetsage2", "display_name": "SheetSage2 (audio to ABC)", "family": "sheetsage2", "path": "models/SheetSage2-GGUF/sheetsage2-orig.gguf", "task": "midi", "mode": "offline", "download_id": "sheetsage2_orig", "min_vram_gb": 8,
|
||
"input_hint_en": "**SheetSage2**: upload a song or instrumental recording to transcribe it into an ABC score artifact." },
|
||
{ "id": "sherpa-kws", "display_name": "Sherpa Zipformer KWS (Chinese / English)", "family": "sherpa_kws", "path": "models/Sherpa-KWS-GGUF/sherpa-kws-zipformer-zh-en-3m-f32.gguf", "task": "wake", "mode": "offline", "download_id": "sherpa_kws_zipformer_zh_en_3m_f32" },
|
||
{ "id": "sidon", "display_name": "Sidon", "family": "sidon", "path": "models/Sidon-GGUF/sidon-v0.1-f32.gguf", "task": "s2s", "mode": "offline", "download_id": "sidon_f32", "workflow": "enhancement" },
|
||
{ "id": "silero-vad", "display_name": "Silero VAD (vad, bundled)", "family": "silero_vad", "path": "assets/framework/models/silero_vad", "task": "vad", "mode": "offline", "min_vram_gb": 1 },
|
||
{ "id": "smart-turn", "display_name": "Smart Turn v3.2", "family": "smart_turn", "path": "models/Smart-Turn-v3-GGUF/smart-turn-v3.2-f32.gguf", "task": "turn", "mode": "offline", "download_id": "smart_turn_v3_2_f32", "workflow": "analysis" },
|
||
{ "id": "soprano-tts", "display_name": "Soprano TTS (tts)", "family": "soprano_tts", "path": "models/Soprano-1.1-80M-GGUF", "task": "tts", "mode": "offline", "download_id": "soprano_1_1_80m_q8_0", "min_vram_gb": 1 },
|
||
{ "id": "sopro-tts", "display_name": "Sopro V2 Turbo (tts + clone)", "family": "sopro_tts", "path": "models/sopro-v2-turbo-GGUF/sopro-v2-turbo-f16.gguf", "task": "tts", "mode": "offline", "download_id": "sopro_v2_turbo_f16", "min_vram_gb": 2, "request_options": ["language", "temperature", "top_p", "top_k", "num_inference_steps", "max_seconds", "min_seconds", "ref_seconds", "text_chunk_size", "seed"] },
|
||
{ "id": "sortformer-diar", "display_name": "Sortformer Diarization 4spk (diar)", "family": "sortformer_diar", "path": "models/diar_sortformer_4spk-v1", "task": "diar", "mode": "offline", "download_id": "sortformer_diar_4spk_v1", "min_vram_gb": 2 },
|
||
{ "id": "stable-audio-medium", "display_name": "Stable Audio 3 Medium (gen)", "family": "stable_audio", "path": "models/stable-audio-3-medium", "task": "gen", "mode": "offline", "download_id": "stable_audio_3_medium", "session_options": { "stable_audio.mem_saver": "true" }, "min_vram_gb": 10 },
|
||
{ "id": "stable-audio-small-music","display_name": "Stable Audio 3 Small Music (gen)", "family": "stable_audio", "path": "models/stable-audio-3-small-music", "task": "gen", "mode": "offline", "download_id": "stable_audio_3_small_music", "min_vram_gb": 4 },
|
||
{ "id": "stable-audio-small-sfx", "display_name": "Stable Audio 3 Small SFX (gen)", "family": "stable_audio", "path": "models/stable-audio-3-small-sfx", "task": "gen", "mode": "offline", "download_id": "stable_audio_3_small_sfx", "min_vram_gb": 4 },
|
||
{ "id": "supertonic", "display_name": "Supertonic 3 (tts 预置音色/多语种)", "display_name_en": "Supertonic 3 (tts, preset voices)", "family": "supertonic", "path": "models/supertonic-3", "task": "tts", "mode": "offline", "download_id": "supertonic_3", "min_vram_gb": 2 },
|
||
{ "id": "tf-gridnet", "display_name": "TF-GridNet (two-speaker separation)", "family": "tf_gridnet", "path": "models/TF-GridNet-GGUF/tf-gridnet-wsj0-2mix-f32.gguf", "task": "sep", "mode": "offline", "download_id": "tf_gridnet_wsj0_2mix_f32" },
|
||
{ "id": "tone-color-vc", "display_name": "Tone Color VC", "family": "tone_color_vc", "task": "vc", "mode": "offline", "path": "models/Tone-Color-VC-GGUF/tone-color-vc-f32.gguf", "download_id": "tone_color_vc_f32" },
|
||
{ "id": "universr-audio", "display_name": "UniverSR Audio", "family": "universr", "path": "models/UniverSR-GGUF", "task": "s2s", "mode": "offline", "download_id": "universr_audio_orig" },
|
||
{ "id": "universr-speech", "display_name": "UniverSR Speech", "family": "universr", "path": "models/UniverSR-GGUF", "task": "s2s", "mode": "offline", "download_id": "universr_speech_orig" },
|
||
{ "id": "vevo2-s2s", "display_name": "Vevo2 (s2s 语音编辑, GGUF Q8)", "display_name_en": "Vevo2 (speech editing, GGUF Q8)", "family": "vevo2", "path": "models/Vevo2-GGUF", "task": "s2s", "mode": "offline", "download_id": "vevo2_gguf", "min_vram_gb": 6,
|
||
"input_hint": "**Vevo2 语音编辑 (s2s)**：上传要编辑的源语音，并在『其它参数(JSON)』里填 `{\"target_text\": \"替换后的完整句子\"}`（编辑保持原说话人音色，可不上传目标音色）。", "input_hint_en": "**Vevo2 speech editing**: upload source speech and set target_text to the complete replacement sentence in Additional options. Editing preserves the original speaker and does not require a target-voice reference." },
|
||
{ "id": "vevo2-svc", "display_name": "Vevo2 (svc 歌声转换, GGUF Q8)", "display_name_en": "Vevo2 (singing voice conversion, GGUF Q8)", "family": "vevo2", "path": "models/Vevo2-GGUF", "task": "svc", "mode": "offline", "download_id": "vevo2_gguf", "min_vram_gb": 6,
|
||
"input_hint": "**Vevo2 歌声转换 (svc)**：上传源歌声 + 目标歌手参考音色，默认 route=style_preserved_svc。style_converted_svc / singing_style_conversion 等风格转换 route 需在『其它参数(JSON)』里补 `style_ref`（服务器本地 wav 路径）/ `style_ref_text` / `target_text`。", "input_hint_en": "**Vevo2 singing conversion**: upload source singing and a target-singer reference. The default route is style_preserved_svc; style-conversion routes also accept style_ref, style_ref_text and target_text in Additional options." },
|
||
{ "id": "vevo2", "display_name": "Vevo2 (vc 语音转换, GGUF Q8)", "display_name_en": "Vevo2 (voice conversion, GGUF Q8)", "family": "vevo2", "path": "models/Vevo2-GGUF", "task": "vc", "mode": "offline", "download_id": "vevo2_gguf", "min_vram_gb": 6 },
|
||
{ "id": "vibevoice", "display_name": "VibeVoice 1.5B/7B (tts, long-form/multi-speaker)", "family": "vibevoice", "path": "models/VibeVoice-1.5B", "task": "tts", "mode": "offline", "download_id": "vibevoice_1_5b", "min_vram_gb": 7 },
|
||
{ "id": "vibevoice-7b", "display_name": "VibeVoice 7B (tts, long-form/multi-speaker)", "family": "vibevoice", "path": "models/VibeVoice-7B-GGUF", "task": "tts", "mode": "offline", "download_id": "vibevoice_7b_q8_0", "min_vram_gb": 16 },
|
||
{ "id": "vibevoice-asr", "display_name": "VibeVoice ASR (asr, 多语种+说话人分段)", "display_name_en": "VibeVoice ASR (asr, multilingual + speaker turns)", "family": "vibevoice_asr", "path": "models/VibeVoice-ASR", "task": "asr", "mode": "offline", "download_id": "vibevoice_asr", "min_vram_gb": 20,
|
||
"input_hint": "**VibeVoice ASR**：自动语种，可输出分段/说话人轮次；文本框可填上下文提示（如 The recording is a meeting conversation.）。权重 17.3G，8G 卡跑不动。", "input_hint_en": "**VibeVoice ASR**: automatic language detection with segment and speaker-turn output. The text box accepts a context prompt. Its 17.3 GB weights require substantially more than 8 GB VRAM." },
|
||
{ "id": "vibevoice-asr-streaming-1.5b", "display_name": "VibeVoice ASR Streaming 1.5B (asr, 多语种+流式)", "display_name_en": "VibeVoice ASR Streaming 1.5B (asr, multilingual + streaming)", "family": "vibevoice_asr_streaming", "path": "models/VibeVoice-ASR-Streaming-1.5B-GGUF/vibevoice-asr-streaming-1.5b-q8_0.gguf", "task": "asr", "mode": "offline", "download_id": "vibevoice_asr_streaming_1_5b_q8_0", "min_vram_gb": 6,
|
||
"input_hint": "**VibeVoice ASR Streaming 1.5B**：7B 的较小版本，功能相同，显存占用更低，准确率略低；文本框可填上下文提示。", "input_hint_en": "**VibeVoice ASR Streaming 1.5B**: the smaller sibling of the 7B with the same features, lower VRAM use, and somewhat lower accuracy. The text box accepts a context prompt." },
|
||
{ "id": "vibevoice-asr-streaming-7b", "display_name": "VibeVoice ASR Streaming 7B (asr, 多语种+流式)", "display_name_en": "VibeVoice ASR Streaming 7B (asr, multilingual + streaming)", "family": "vibevoice_asr_streaming", "path": "models/VibeVoice-ASR-Streaming-7B-GGUF/vibevoice-asr-streaming-7b-q8_0.gguf", "task": "asr", "mode": "offline", "download_id": "vibevoice_asr_streaming_7b_q8_0", "min_vram_gb": 18,
|
||
"input_hint": "**VibeVoice ASR Streaming 7B**：多语种 ASR，支持长音频流式解码和说话人轮次；文本框可填上下文提示。", "input_hint_en": "**VibeVoice ASR Streaming 7B**: multilingual ASR with long-audio streaming decode and speaker-turn output. The text box accepts a context prompt." },
|
||
{ "id": "vieneu-tts", "display_name": "VieNeu-TTS v3 Turbo (tts 越南语, 社区)", "display_name_en": "VieNeu-TTS v3 Turbo (vi tts, community)", "family": "vieneu_v3_turbo", "path": "models/VieNeu-TTS-v3-Turbo", "task": "tts", "mode": "offline", "download_id": "vieneu_v3_turbo_v3_turbo", "min_vram_gb": 4,
|
||
"input_hint": "**VieNeu-TTS v3 Turbo**（社区模型）：越南语 / 英语；上传参考音色即克隆。",
|
||
"input_hint_en": "**VieNeu-TTS v3 Turbo** (community): Vietnamese and English. Upload a reference voice to clone." },
|
||
{ "id": "voxcpm1", "display_name": "VoxCPM1 0.5B (tts + clone)", "family": "voxcpm1", "path": "models/VoxCPM1-GGUF", "task": "tts", "mode": "offline", "download_id": "voxcpm1_0_5b_q8_0", "min_vram_gb": 4 },
|
||
{ "id": "voxcpm2", "display_name": "VoxCPM2 (tts)", "family": "voxcpm2", "path": "models/VoxCPM2", "task": "tts", "mode": "offline", "download_id": "voxcpm2", "session_options": { "voxcpm2.weight_type": "q8_0" }, "min_vram_gb": 6 },
|
||
{ "id": "voxtral-realtime", "display_name": "Voxtral Mini 4B Realtime (asr, 自动语种+流式)", "display_name_en": "Voxtral Mini 4B Realtime (asr, auto + streaming)", "family": "voxtral_realtime", "path": "models/Voxtral-Mini-4B-Realtime-2602-GGUF", "task": "asr", "mode": "offline", "download_id": "voxtral_realtime", "min_vram_gb": 8 },
|
||
{ "id": "voxtral-tts", "display_name": "Voxtral 4B TTS", "family": "voxtral_tts", "path": "models/Voxtral-4B-TTS-2603-GGUF/voxtral-4b-tts-2603-bf16.gguf", "task": "tts", "mode": "offline", "download_id": "voxtral_4b_tts_2603_bf16", "default_text": "The next train leaves in ten minutes. Please wait by platform five." },
|
||
{ "id": "whistle-asr", "display_name": "Whistle ASR", "family": "whistle_asr", "path": "models/Whistle-GGUF/whistle-f32.gguf", "task": "asr", "mode": "offline", "download_id": "whistle_asr_f32" },
|
||
{ "id": "yue2", "display_name": "Yue2 3B (song gen)", "family": "yue2", "path": "models/Yue2-3B-GGUF", "task": "gen", "mode": "offline", "download_id": "yue2_main_q8_0", "default_text": "[Verse]\nSoft morning light is touching the window.\nI hear the city waking below.\n[Chorus]\nStay with the rhythm, let it carry us home.\nSing with the sunrise, we are never alone.", "min_vram_gb": 12,
|
||
"input_hint_en": "**Yue2 3B**: provide lyrics and a style prompt. The default inference combo is Main Q8_0 + VAE F16; choose other main/VAE files from Model parameters before loading." },
|
||
{ "id": "zipenhancer", "display_name": "ZipEnhancer", "family": "builtin_audio_utils", "task": "s2s", "workflow": "enhancement", "mode": "offline", "path": "models/Audio-Utilities/zipenhancer/zipenhancer.safetensors", "load_options": { "utility": "zipenhancer" }, "weights_url": "https://raw.githubusercontent.com/0xShug0/audio.cpp/main/assets/framework/audio_utilities/zipenhancer/zipenhancer.safetensors" },
|
||
{ "id": "zipvoice", "display_name": "ZipVoice (中英零样本克隆)", "display_name_en": "ZipVoice (zh/en zero-shot cloning)", "family": "zipvoice", "path": "models/ZipVoice-Distill-GGUF", "task": "clon", "mode": "offline", "download_id": "zipvoice_distill_gguf", "min_vram_gb": 4,
|
||
"input_hint": "**ZipVoice**：中/英零样本声音克隆（k2-fsa TTSZipformer flow-matching + Vocos，蒸馏版 8 步采样）。上传参考音频，并在『参考文本’里填入它的逐字转写——参考音频与参考文本必须配对；合成文本支持中文、英文及混排。权重 Apache-2.0。",
|
||
"input_hint_en": "**ZipVoice**: zero-shot voice cloning in zh/en (k2-fsa TTSZipformer flow matching + Vocos, distilled 8-step sampling). Upload a reference clip and put its exact transcript in the reference-text box — the clip and the transcript must match; synthesis text supports zh/en/mixed. Weights are Apache-2.0." }
|
||
]
|
||
}
|