Ant Group open-sources Ming-3.0: five speech models spanning TTS, ASR and audio understanding up to 126B
AccBalanced · x · 2026-10-20
Ant Group's AntLingAGI open-sourced five Ming-3.0 speech models across three families:
- Ming-3.0-tts: 0.5B for voice design & paralinguistic cues; 7.9B (A1.3B) for long-form speech
- Ming-3.0-audio: Flash 126B (A6.5B) for understanding & tool use; Tiny 9.3B (A2.7B) for efficient dialogue
- Ming-3.0-asr: Tiny 8.6B (A2B) supporting dialects and context
vLLM-Omni ships Day-0 support, so all models can be served immediately.
Related event: Ant's inclusionAI Open-Sources Five Ming-3.0 Speech Models(6 posts)→
More from Models
- Cloudflare launches Clairic, a new extraction model class claiming higher accuracy and lower cost than LLMs — ritakozlov · 2026-10-20
- Users want open-format export from Claude Docs as charts only save as images — BlackHC · 2026-10-20
- EMNLP study: LLM safety interventions drop up to 4.5x when distress is wrapped in delusion — MaartenSap · 2026-10-20
- Cybersecurity-focused VLM OrcaCyber-Nano-1.0 trends on Hugging Face — orcarouter · 2026-10-20
- 0.8B local model with 2,000 labels beats Jev's frontier API on banking data — kalyan_kpl · 2026-10-20
- New TTS benchmark tests 26 models across 30,564 requests: fastest at 46ms — mattturck · 2026-10-20