mlx-community/Ming-3.0-audio-tiny-bf16

Ming-3.0 Audio Tiny for MLX

Converted from inclusionAI/Ming-3.0-audio-tiny.

English and Chinese audio understanding, transcription, text chat and spoken dialogue. task='spoken_dialogue' loads the bundled Talker lazily and produces 44.1 kHz audio. The Thinker, audio encoder, Talker and audio VAE use BF16 weights. Speculative MTP decoding and streaming are not implemented. Reference speaker extraction uses the bundled native MLX CAMPPlus encoder. Its weights retain FP32 precision. No ONNX packages are required.

from mlx_audio.sts import load
model = load('mlx-community/Ming-3.0-audio-tiny-bf16', strict=True)
result = model.generate("speech.wav", task="audio_understanding",
                        text="Describe what you hear.")
print(result.text)

Offline inference on Apple Silicon using native MLX code. The tokenizer is loaded without executing remote code.

Source checkpoint revision: 7cbe4a75c979f6ea48f1ee31998eb02bd7012233.

This export includes all weights and tokenizers needed for its supported tasks. The native CAMPPlus weights are bundled under talker/speaker_encoder; their Apache-2.0 license and attribution accompany the weights. The complete Talker, its text tokenizer and audio VAE are included under talker/. No separate model downloads or ONNX packages are required. Use an MLX Audio version with Ming Tiny support.

The example repository name above may be replaced with the local export directory or your uploaded repository ID.

Validation: strictly reloaded from a relocated directory with an empty Hugging Face cache, network access blocked and ONNX packages absent. See validation.json for the tested tasks and weight inventory. SHA256SUMS covers all package files.