←back to thread

Qwen3-Omni-Flash-2025-12-01：a next-generation native multimodal large model

(qwen.ai)
314 points pretext | 5 comments | 10 Dec 25 16:13 UTC | HN request time: 0.682s | source
1. aschobel ◴[10 Dec 25 18:03 UTC] No.46221096[source]▶
Looks to be API only. Bummer.
replies(3): >>46221222 #>>46222976 #>>46262907 #
2. ◴[10 Dec 25 18:12 UTC] No.46221222[source]▶
3. readyplayeremma ◴[10 Dec 25 20:05 UTC] No.46222976[source]▶
The models are right here, one of the first links in the post: https://huggingface.co/collections/Qwen/qwen3-omni

edit: Nevermind, in spite of them linking it at the top, they are the old models. Also, the HF demo is calling their API and not using HF for compute.

replies(1): >>46224501 #
4. aschobel ◴[10 Dec 25 22:00 UTC] No.46224501[source]▶
It is super confusing. I also thought this initially was open weights.
5. Alifatisk ◴[14 Dec 25 13:34 UTC] No.46262907[source]▶
It seems to be available on Qwen chat? https://chat.qwen.ai/settings/model?id=qwen3-omni-flash-2025...