Video wird geladen...

Video konnte nicht geladen werden

Zur Startseite

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

119,953 Aufrufe • vor 4 Monaten •via X (Twitter)

32 Kommentare

Profilbild von AK
AKvor 4 Monaten

paper:

Profilbild von Sanjay
Sanjayvor 4 Monaten

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

Profilbild von Progyan
Progyanvor 4 Monaten

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

Profilbild von Pils10
Pils10vor 4 Monaten

AI Wifus coming soon!

Profilbild von Adam
Adamvor 4 Monaten

that's really cool

Profilbild von Völkisch
Völkischvor 4 Monaten

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.

Profilbild von James
Jamesvor 4 Monaten

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

Profilbild von briton
britonvor 4 Monaten

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

Profilbild von Bin Chen
Bin Chenvor 4 Monaten

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.

Profilbild von James' AI Takes
James' AI Takesvor 4 Monaten

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.

Profilbild von Hrittik Das
Hrittik Dasvor 4 Monaten

she was more expressive than the guy

Profilbild von Joannot Fampionona
Joannot Fampiononavor 4 Monaten

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

Profilbild von Hammad Hamid
Hammad Hamidvor 4 Monaten

amazing man

Profilbild von ΛIΚO
ΛIΚOvor 4 Monaten

@grok wtf is this real?

Profilbild von Gordon Olson
Gordon Olsonvor 4 Monaten

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

Profilbild von ansuman
ansumanvor 4 Monaten

Final piece in AI wifu!

Profilbild von Colbert
Colbertvor 4 Monaten

I like this

Profilbild von oo
oovor 4 Monaten

is this available to use anywhere?

Profilbild von TongHe
TongHevor 4 Monaten

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.

Profilbild von Zakaria
Zakariavor 4 Monaten

@Justiniansli

Profilbild von JackWison
JackWisonvor 4 Monaten

good idear

Profilbild von Cris Lenta
Cris Lentavor 4 Monaten

so awesome!

Profilbild von sarahgamedev
sarahgamedevvor 4 Monaten

Oh gawd, I am real. She’s not real

Profilbild von makyō
makyōvor 4 Monaten

Model/code release?

Profilbild von ☣ GBLW 
☣ GBLW vor 4 Monaten

Great !

Profilbild von TechMD
TechMDvor 4 Monaten

amazing work

Profilbild von veloX
veloXvor 4 Monaten

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.

Profilbild von MinG.AI
MinG.AIvor 4 Monaten

This so unreal brooo

Profilbild von Sebastian Buzdugan
Sebastian Buzduganvor 4 Monaten

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

Profilbild von Taran Mann
Taran Mannvor 4 Monaten

Is it on Comfy yet

Profilbild von Raj
Rajvor 4 Monaten

Great

Profilbild von ReggieJ
ReggieJvor 4 Monaten

我要做一个 @CuiMao 的

Ähnliche Videos

Time to end it
5:29

Sensitive content

Time to end it

NRM84

22,902 Aufrufe • vor 1 Jahr