{"version":1,"type":"story","url":"https://digestai.news/story/inclusionai-releases-ming-3-0-audio-flash-a-multilingual-audio-languag","json":"https://digestai.news/story/inclusionai-releases-ming-3-0-audio-flash-a-multilingual-audio-languag.json","markdown":"https://digestai.news/story/inclusionai-releases-ming-3-0-audio-flash-a-multilingual-audio-languag.md","slug":"inclusionai-releases-ming-3-0-audio-flash-a-multilingual-audio-languag","headline":"InclusionAI releases Ming-3.0-audio-flash, a multilingual audio-language model","summary":"inclusionAI has open‑sourced ming-3.0-audio-flash, a native audio‑language model that processes spoken input together with text instructions. The model handles Chinese and English speech, including dialects, accents, noisy environments and multi‑speaker recordings, and can perform long‑form audio understanding, acoustic scene analysis, and reasoning over audio content. It also offers optional speech synthesis through a built‑in Talker that can use preset voices or reference‑audio prompts.\n\nThe repository provides a Python example that runs the model with the Transformers library, as well as deployment scripts for vLLM‑Omni (using eight 80 GB H800 GPUs for a 128K serving profile) and SGLang (four GPUs for text‑only output). The project is released under the MIT License, and the GitHub page recorded 22 downloads in the last month. inclusionAI notes that users should review the code before enabling remote execution flags, reflecting a focus on security and trust.\n\nThe release adds a new multimodal capability to the Hugging Face ecosystem, allowing developers to build applications that understand and respond to audio in multiple languages without needing separate speech‑to‑text and language models.","keyPoints":["inclusionAI released ming-3.0-audio-flash, supporting speech recognition, audio reasoning, and optional speech synthesis for Chinese and English.","Model can be deployed with vllm‑omni on eight 80 GB H800 GPUs for 128K context or with sglang on four GPUs for text output.","The GitHub repository logged 22 downloads last month and is distributed under the MIT License."],"whyItMatters":"A single model that jointly processes audio and language lowers integration complexity and enables new applications that can understand and answer spoken queries across languages.","category":{"slug":"models","name":"Generative AI & Models","url":"https://digestai.news/category/models"},"entities":{"companies":["inclusionAI"],"models":["Ming-3.0-audio-flash"],"people":[]},"firstPublishedAt":"2026-10-19T14:27:16Z","updatedAt":"2026-10-19T14:29:36Z","sourceCount":2,"hasPrimarySource":true,"sources":[{"outlet":"huggingface.co","title":"inclusionAI/Ming-3.0-audio-flash · Hugging Face","url":"https://huggingface.co/inclusionAI/Ming-3.0-audio-flash","publishedAt":"2026-10-19T14:27:16Z","type":"primary","primary":true,"lead":true},{"outlet":"huggingface.co","title":"inclusionAI/Ming-3.0-audio-tiny · Hugging Face","url":"https://huggingface.co/inclusionAI/Ming-3.0-audio-tiny","publishedAt":"2026-10-19T14:29:36Z","type":"primary","primary":true,"lead":false}],"sourceNotes":null,"discussions":[{"site":"Reddit","url":"https://www.reddit.com/r/LocalLLaMA/comments/1x9yf3m/inclusionaiming30audiotiny_hugging_face/","points":null},{"site":"Reddit","url":"https://www.reddit.com/r/LocalLLaMA/comments/1x9yd01/inclusionaiming30audioflash_hugging_face/","points":null}],"thread":null,"cite":{"text":"Digest AI, \"InclusionAI releases Ming-3.0-audio-flash, a multilingual audio-language model\", 19 October 2026, https://digestai.news/story/inclusionai-releases-ming-3-0-audio-flash-a-multilingual-audio-languag","publisher":"Digest AI","title":"InclusionAI releases Ming-3.0-audio-flash, a multilingual audio-language model","datePublished":"2026-10-19T14:27:16Z","url":"https://digestai.news/story/inclusionai-releases-ming-3-0-audio-flash-a-multilingual-audio-languag"},"generatedBy":"Written by Digest AI's editorial model from the linked sources; the sources are the record.","license":"Headlines, digests and key points are written by Digest AI and may be quoted with a link to the story page. Linked articles belong to their publishers. Terms: https://digestai.news/terms#reuse"}