k2-fsa/OmniVoice
Full fine-tune vs Qwen/Qwen3-0.6B. Fine-tune: 100% of the weights changed, across 28 of 28 layers.
Tensors changed: Norms 113/113, MLP 84/84, Attention 112/112
27 derivatives checked
- Serveurperso/OmniVoice-GGUF: Same weights, converted
- cstr/omnivoice-GGUF: Not comparable yet
- mohammedaly22/VoiceTut-TTS: Full fine-tune
- mlx-community/OmniVoice-bfloat16: Same weights, converted
- splendor1811/omnivoice-vietnamese: Full fine-tune
- kjanh/KhanhTTS-OmniVoice: Full fine-tune
- hotdogs/omnivoice-thai: Full fine-tune
- drbaph/OmniVoice-bf16: Same weights, converted
- mlx-community/OmniVoice-8bit: Quantization
- oddadmix/lahgtna-omnivoice-v2: Full fine-tune
- ModelsLab/omnivoice-singing: Full fine-tune
- Supbatomic/OmniVoice-bf16: Same weights, converted
- Lab-et/omnivoice-amharic: Full fine-tune
- gheero-Leyu/amharic-omnivoice-tts: Full fine-tune
- mlx-community/OmniVoice-4bit: Quantization
- fablevi/OmniVoice-GGUF-clone: Same weights, converted
- ai-ssam/darija-omnivoice-kore-v1: Full fine-tune
- CreativeHub008/OmniVoice-GGUF: Same weights, converted
- SwinliQ-AIn/OmniVoice-MLX-4bit: Quantization
- mlx-community/OmniVoice: Same weights, repacked
- edwixx/omnivoice-brpt-v15: Full fine-tune
- ldov/OmniVoice-GGUF: Same weights, converted
- kristijonas/krapas-Regina: Full fine-tune
- NMikka/omnivoice-finetuned-ka: Full fine-tune
- remember2015/omnivoice-mlx-q8-fp16: Quantization
- Supbatomic/OmniVoice-GGUF: Same weights, converted
- Bonenk/OmniVoice-GGUF: Not comparable yet
68 other repos hold byte-for-byte identical weight files.