vanch007/Ming-3.0-tts-tiny-mlx-4bit

Pure Apple Silicon MLX native port of inclusionAI/Ming-3.0-tts-tiny (7.9B MoE with 128 Experts + 1 Shared Expert + KDA Attention) in 4-bit Quantized (Ultra-Lightweight).

Architecture: 128 Routed Experts + 1 Shared Expert + KDA Recurrent Linear Attention
Size: 5.42 GB
Zero Triton: Pure Metal acceleration on Apple Silicon (M1/M2/M3/M4)

Sample Audios

Task	Text	Audio
Podcast	speaker_1: 你最近是不是又去那家面馆了？ speaker_2: 对啊，排了半个小时的队，不过真的值得。	Audio
Lingxiaomeng	夜晚的微风吹散了白天的喧嚣，愿好梦常伴你左右。	Audio
Text Norm	订单号 20250317，共 3 件商品，合计 ¥1,280.50，预计 3-5 天送达。	Audio
Fast Speed	今晚的观测将在云层散开后开始。	Audio
Sad Emotion	所有的测试都通过了。	Audio

Downloads last month: 19

MLX

Hardware compatibility

Quantized

Model tree for vanch007/Ming-3.0-tts-tiny-mlx-4bit

Base model

inclusionAI/Ming-3.0-tts-tiny

Quantized

(6)

this model