Live and recent posts from across the Fediverse tagged #qwen3omni, aggregated by home.social.
🎧 Multimodal ingest covers speech, speaker identity, sound events, multi-speaker conversations and music, bundling ASR, voiceprint, scene and emotion detection
🤖 Model family with #Qwen25Omni, #Qwen3Omni and #StepAudio2Mini, trained on the open ChatMem-400K dataset through a three-stage OPD pipeline, finetuning code included
Now that dangerously-powerful #GenAI is out in the wild, running locally to create audio, video, text, images - like with #qwen3omni
I'm counting the days until "Copyright as we know it" will turn into "humankind will figure out, that it's over"
...and we can finally live freely:
dealing with the fact that we can /copy anything/.
At computing cost.
Please:
#ValueTheCreators and #artists and use this to get *free time* to go out and engage with real people.
Be happy again. 😄
Wirtschaftlicher Impuls 💼 Durch #offeneLizenz und #starkePerformance könnte #Alibaba die Dominanz westlicher #KIModelle erstmals ernsthaft herausfordern.
Neue Ära der KI 🔮 #Qwen3Omni zeigt, wie offene, leistungsstarke #Multimodal-Modelle den Zugang zu #KI demokratisieren und Innovation beschleunigen.
👉 https://eicker.TV ▹ #Technik #Medien #Politik #Wirtschaft ⏻ https://eicker.BE/ratung von Gerrit Eicker aus Münster (2/2)
Alibabas KI-Modell versteht Text, Bild, Audio und Video als Eingabeinstrument.
Alibaba legt vor 🚀 Mit #Qwen3Omni präsentiert #Alibaba ein #OpenSource-#KI-Modell, das Text, Bild, Audio und Video versteht – und Text sowie Sprache ausgeben kann.
Omni-Integration 🎥 Im Gegensatz zu Konkurrenzmodellen ist Qwen3-Omni nativ #multimodal entwickelt und ermöglicht schnelle, Echtzeit-Interaktionen. (1/2)
#Alibaba’s #Qwen3Omni, an #opensource #AImodel, accepts text, audio, image, and video inputs, and outputs text and audio. It outperforms proprietary models like OpenAI’s GPT-4o and Google’s Gemini 2.5 Pro in various benchmarks, including text, reasoning, speech, audio, image, and video tasks. #Apache 2.0 licence, #free download, #commercialuse. https://venturebeat.com/ai/chinas-alibaba-challenges-u-s-tech-giants-with-open-source-qwen3-omni-ai?eicker.news #tech #media #news
🎉 Behold, Qwen3-Omni: the Swiss Army knife of AI, capable of doing everything except making your morning coffee. Alibaba Cloud's new toy can process text, audio, images, and video, but good luck figuring out which button to press first. 🤖📸🔊
https://github.com/QwenLM/Qwen3-Omni #Qwen3Omni #AlibabaCloud #AItechnology #Innovation #TechTrends #HackerNews #ngated
Qwen3-Omni: Native Omni AI Model for Text, Image & Video
https://github.com/QwenLM/Qwen3-Omni
#HackerNews #Qwen3Omni #OmniAI #TextImageVideo #AIModel #GitHub