Proudly present PAW-27B-X3, 3.000 people download this from huggingface.
Compressed model from Qwen 3.8 27B at 3.5bpw only 11.5GiB fit on single GPU RTX 3090 at full context 256k. Run test with Oh My Pi agent harness to produce single html Eiffel Tower.
Et h a b e m u s ... C i n e m a g r a m !!
First tests !! CompfyUI + ChatGPT6 (make that json for ...) + MBP M3 max 128 GB Next : add sound, then voice, then lipssync, then improve quality, then regaularity (scene, characters...). Then add night work (8h...), then
There's now an AI that watches a two-hour video the way you would.
It sees the screen. It hears every word. And it never loses the plot. It's Alibaba's Qwen 3.8 Omni Flash — and the clever part isn't that it holds more. → It doesn't chew through every second of a video →
Alibaba just quietly shipped an AI that watches video like a person — start to finish.
No transcripts. No screenshots. No doing half the work yourself. It's called Qwen 3.8 Omni Flash. The word that matters is "omni." Most models read text. Some can look at a picture.
You have reached the end of the archive
All of qwen38