Загрузка видео...

Не удалось загрузить видео

На главную

Wan-Streamer v0.1 End-to-end Real-time Interactive Foundation Models

119,953 просмотров • 4 месяцев назад •via X (Twitter)

Комментарии: 32

Фото профиля AK
AK4 месяцев назад

paper:

Фото профиля Sanjay
Sanjay4 месяцев назад

Single guys gonna love it for sure and gold diggers rn:

Фото профиля Progyan
Progyan4 месяцев назад

It's almost opensource if someone is willing to contribute an extra 10% of perf.

Фото профиля Pils10
Pils104 месяцев назад

AI Wifus coming soon!

Фото профиля Adam
Adam4 месяцев назад

that's really cool

Фото профиля Völkisch
Völkisch4 месяцев назад

That's very impressive, its expressions are very realistic. I hope they release the models for us to play around with. Perhaps we can add custom characters to it via training.

Фото профиля James
James4 месяцев назад

need full upper body like at desk or sitting etc so you can see the hand gestures not cut off, so can be composited on scene

Фото профиля briton
briton4 месяцев назад

another stunning example of Alibaba's new model - Ali-Catfish1.0

Фото профиля Bin Chen
Bin Chen4 месяцев назад

avtr-1 is doing this pretty good with a single 3090 in realtime, it renders lip-synced speech and active listening at 25 fps on a single GPU. the only drawback is it only has facial change but not neck rotation etc.

Фото профиля James' AI Takes
James' AI Takes4 месяцев назад

real-time interactive foundation models are the fun part. the ugly part is whether latency, state, controllability, and failure recovery survive past the demo clip. streaming magic still needs boring systems engineering.

Фото профиля Hrittik Das
Hrittik Das4 месяцев назад

she was more expressive than the guy

Фото профиля Joannot Fampionona
Joannot Fampionona4 месяцев назад

We need that ASAP in the wan app even if it's a paid feature

Фото профиля Hammad Hamid
Hammad Hamid4 месяцев назад

amazing man

Фото профиля ΛIΚO
ΛIΚO4 месяцев назад

@grok wtf is this real?

Фото профиля Gordon Olson
Gordon Olson4 месяцев назад

This is really cool. Was thinking about this last week. But didn’t take it as far as full transformer architecture for everything. So this is fascinating I’ll be looking at this to run via webgpu

Фото профиля ansuman
ansuman4 месяцев назад

Final piece in AI wifu!

Фото профиля Colbert
Colbert4 месяцев назад

I like this

Фото профиля oo
oo4 месяцев назад

is this available to use anywhere?

Фото профиля TongHe
TongHe4 месяцев назад

Real-time interactive AI is impressive. I think the next hard problem is shared-room conversation: multiple people, multiple languages, interruptions, accents, names, and context. Not just one human and one AI. A whole group staying in the same conversation.

Фото профиля Zakaria
Zakaria4 месяцев назад

@Justiniansli

Фото профиля JackWison
JackWison4 месяцев назад

good idear

Фото профиля Cris Lenta
Cris Lenta4 месяцев назад

so awesome!

Фото профиля sarahgamedev
sarahgamedev4 месяцев назад

Oh gawd, I am real. She’s not real

Фото профиля makyō
makyō4 месяцев назад

Model/code release?

Фото профиля ☣ GBLW 
☣ GBLW 4 месяцев назад

Great !

Фото профиля TechMD
TechMD4 месяцев назад

amazing work

Фото профиля veloX
veloX4 месяцев назад

Gerçek zamanlı etkileşim için video akışı yeterli değil — sistem lokal girdi işleyebiliyorsa neden AI ajanın web kameraya erişmesi gerekiyor? Latency, gizlilik ve inference maliyeti açısından stream-based mimari daha pahalı.

Фото профиля MinG.AI
MinG.AI4 месяцев назад

This so unreal brooo

Фото профиля Sebastian Buzdugan
Sebastian Buzdugan4 месяцев назад

once end to end delay passes 200ms interactivity feels fake can wan-streamer stay below that

Фото профиля Taran Mann
Taran Mann4 месяцев назад

Is it on Comfy yet

Фото профиля Raj
Raj4 месяцев назад

Great

Фото профиля ReggieJ
ReggieJ4 месяцев назад

我要做一个 @CuiMao 的

Похожие видео

Time to end it
5:29

Sensitive content

Time to end it

NRM84

22,902 просмотров • 1 год назад