One image + text + camera trajectory = controllable worlds. All on a single GPU.
93,607 次观看
33 条评论
评论来自 X (Twitter) 上的原帖

@NVIDIAAI
The architecture features Hybrid Linear Attention, Dual-Branch Camera Control, Two-Stage Generation Pipeline, and a Robust Annotation Pipeline. All of this combines for stronger action-following accuracy and higher throughput while maintaining visual quality.
48 个赞

@NVIDIAAI
You can find the full project including the paper, code, and model here 👇
47 个赞

@PublicAI_
this sounds super cool! creating worlds with just a single gpu is a game changer. can't wait to see what people do with SANA-WM!
3 个赞

@TheAIColonyRD
60 seconds of video. One image to start. One GPU to run it. Open source. The game engine just became a prompt.
2 个赞

@usepapertrace
Looks great. Can you link the paper and repo, name training data sources/licenses, and state the exact GPU/VRAM for the single GPU claim? That would let others reproduce the camera control results.
1 个赞

@kekkodamato_
2.6B for 60-second world-consistent video with camera control is a wild benchmark. The open-source release is what makes this actually useful — most teams don't need Hollywood quality, they need something they can run and fine-tune. What hardware are people actually running inference on?
1 个赞

@heynaavi
love how you're pushing the boundaries of video generation with SANA-WM
1 个赞

@iamsupersocks
@llmgram
1 个赞

@thaonlyjonathan
You're on the right track Nvidia, but please make it more performant and keep training. Go full send on compute for training.
1 个赞

@ashwathama
SANA-WM 2.6B open source world model for 60-second video on a single GPU
1 个赞

@CreativeSideB
Looking very soulless, generic and not worth it!
1 个赞

@_Ajeya
Can it run on a 4090?
1 个赞

@drayinvests
$NVDA seriously is going to rule the world
1 个赞

@ApexMetaX
One image + text + trajectory = controllable worlds on a single GPU. Simulation is becoming cheap and precise.
1 个赞

@chikoevm
bbox queen looks clean, seedance 2 with gpt image 2, huh? got a side by side without seedance to compare?

@timsius
Love this

@asimcodes
"single GPU" is the part that changes everything here. world models for video have always felt like a data center problem, shrinking that down to one GPU opens this up to a completely different audience

@xtaxrich
SANA-WM is concrete enough to test. One image, text, camera path, and single GPU are the setup. Control drift is the receipt.

@SamMusical6990
2.6B for 60-second generation is compelling if SANA-WM keeps temporal drift under control.

@EconEpoch
@KashJim50723 🤷♂️












