Loading video...

Video Failed to Load

Go Home

Sweet. RAVEN - turns MiniMax‑H3 into a real-time video streaming engine. It streams video chunk by chunk like a livestream, no drift, no glitches even on long runs.

wildmindai's profile picture

Wildminder

11,205 subscribers

11,882 views • 15 days ago •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

PixVerse just launched their first real-time video world model R1, powered by fal's high-performance compute. It turns AI video from static clips into a real-time, interactive world. 🔄 Continuous generation stream 🧠 Autoregressive memory system ⚡ Ultra-fast response engine
0:27

PixVerse just launched their first real-time video world model R1, powered by fal's high-performance compute. It turns AI video from static clips into a real-time, interactive world. 🔄 Continuous generation stream 🧠 Autoregressive memory system ⚡ Ultra-fast response engine

fal

19,243 views • 7 months ago

Grok Imagine workflow, simplified for humans Start off by building a static image. Bring it to life by adding motion. Grok turns it into a video like it was obvious - no sweat!
0:38

Grok Imagine workflow, simplified for humans Start off by building a static image. Bring it to life by adding motion. Grok turns it into a video like it was obvious - no sweat!

Mario Nawfal

41,241 views • 7 months ago

Minimax H3 Max has generates video faster than you can watch it so I hooked it to a twitch livestream! Now you can watch infinite interdimensional cable - link to the stream below
0:23

Minimax H3 Max has generates video faster than you can watch it so I hooked it to a twitch livestream! Now you can watch infinite interdimensional cable - link to the stream below

Rehan Sheikh

6,007,095 views • 5 days ago

Introducing AI video you can watch and interact with, in real-time! Powering this is a new world model that imagines and streams video frames every 40ms(!). No game engine in sight. We call it interactive video, and it's free for anyone to try right now (GPUs permitting)!
0:32

Introducing AI video you can watch and interact with, in real-time! Powering this is a new world model that imagines and streams video frames every 40ms(!). No game engine in sight. We call it interactive video, and it's free for anyone to try right now (GPUs permitting)!

Odyssey

705,075 views • 1 year ago

someone open-sourced a tool that turns any phone video into a full 3D world in real time. you record a walk around your house, drop the mp4 in, and the entire scene comes back as a walkable 996k-point world at 60fps on your local GPU. no cloud. no subscription. runs 100% locally. 100% open source.
0:33

someone open-sourced a tool that turns any phone video into a full 3D world in real time. you record a walk around your house, drop the mp4 in, and the entire scene comes back as a walkable 996k-point world at 60fps on your local GPU. no cloud. no subscription. runs 100% locally. 100% open source.

Superman

818,134 views • 1 month ago

I think I’m kinda obsessed with MiniMax Design now One prompt. That’s it. What was I even doing before this? Just a few minutes and one prompt gets you a full retro Japanese travel-poster-style video. MiniMax Design really brings out the best in H3 And no, I’m not exaggerating
0:27

I think I’m kinda obsessed with MiniMax Design now One prompt. That’s it. What was I even doing before this? Just a few minutes and one prompt gets you a full retro Japanese travel-poster-style video. MiniMax Design really brings out the best in H3 And no, I’m not exaggerating

Jaynit Makwana

36,397 views • 15 days ago

Real-time video editing is here. SANA-Streaming by @NVIDIA is now live on Reactor. Change the style, the scene, the mood of any video as it plays, with no rendering wait. Try it now:
0:25

Real-time video editing is here. SANA-Streaming by @NVIDIA is now live on Reactor. Change the style, the scene, the mood of any video as it plays, with no rendering wait. Try it now:

reactor

67,191 views • 2 months ago

Honestly stunned by MiniMax Design (H3) MiniMax H3 Open. I ran the model locally in ComfyUI on a 4090 RTX. Created the full 4-minute “Just NPCs” ft. Crissy video. Video result rivals the best models I’ve tried. Topaz Labs upscale to 4K60. 🎬 Cyberpunk Girl Crissy is Back!
4:10

Honestly stunned by MiniMax Design (H3) MiniMax H3 Open. I ran the model locally in ComfyUI on a 4090 RTX. Created the full 4-minute “Just NPCs” ft. Crissy video. Video result rivals the best models I’ve tried. Topaz Labs upscale to 4K60. 🎬 Cyberpunk Girl Crissy is Back!

OrctonAI

22,663 views • 3 days ago

Today we’re excited to share a peek at our next-gen model. Hope you like it 🥹 Imagine a video model you can play like a game, all in real-time. That’s what we’re building. We call it interactive video, and there's no game engine in sight, just a model’s imagination.
0:30

Today we’re excited to share a peek at our next-gen model. Hope you like it 🥹 Imagine a video model you can play like a game, all in real-time. That’s what we’re building. We call it interactive video, and there's no game engine in sight, just a model’s imagination.

Odyssey

207,705 views • 1 year ago

UE5 AI Motion Plugin — efficiently converts video into skeletal points and facial blendshapes in real time. Powered by a 1-billion-parameter motion model trained on NVIDIA CUDA, it runs locally with 8GB of VRAM and transforms webcam or video footage into real-time XYZ skeletal point data and rotation values.
0:33

UE5 AI Motion Plugin — efficiently converts video into skeletal points and facial blendshapes in real time. Powered by a 1-billion-parameter motion model trained on NVIDIA CUDA, it runs locally with 8GB of VRAM and transforms webcam or video footage into real-time XYZ skeletal point data and rotation values.

CYANPUPPETS

13,751 views • 9 months ago

UE5.6 AI Real-Time Motion Generation Plugin — powered by a 1-billion-parameter motion model trained on NVIDIA CUDA. It runs locally in real time with 8GB of VRAM and converts webcam or video footage into real-time XYZ skeletal point data and rotation values.
1:33

UE5.6 AI Real-Time Motion Generation Plugin — powered by a 1-billion-parameter motion model trained on NVIDIA CUDA. It runs locally in real time with 8GB of VRAM and converts webcam or video footage into real-time XYZ skeletal point data and rotation values.

CYANPUPPETS

28,871 views • 8 months ago

UE5.6 AI Real-Time Motion Generation Plugin — powered by a 1-billion-parameter motion model trained on NVIDIA CUDA. It runs locally in real time with 8GB of VRAM and converts webcam or video footage into real-time XYZ skeletal point data and rotation values.
0:39

UE5.6 AI Real-Time Motion Generation Plugin — powered by a 1-billion-parameter motion model trained on NVIDIA CUDA. It runs locally in real time with 8GB of VRAM and converts webcam or video footage into real-time XYZ skeletal point data and rotation values.

CYANPUPPETS

15,968 views • 7 months ago

UE5.6 AI Real-Time Motion Generation Plugin — powered by a 1-billion-parameter motion model trained on NVIDIA CUDA. It runs locally in real time with 8GB of VRAM and converts webcam or video footage into real-time XYZ skeletal point data and rotation values.
0:23

UE5.6 AI Real-Time Motion Generation Plugin — powered by a 1-billion-parameter motion model trained on NVIDIA CUDA. It runs locally in real time with 8GB of VRAM and converts webcam or video footage into real-time XYZ skeletal point data and rotation values.

CYANPUPPETS

29,029 views • 7 months ago

Nice 😍 You can now long press any image on X and Grok turns it into a video 🤩
0:31

Nice 😍 You can now long press any image on X and Grok turns it into a video 🤩

Mukul Sharma

15,066 views • 1 year ago

Single video in, 4D human out. 4DAnyone turns a casual monocular video into a 4DGS model. No camera rig, no calibration, no tripod. Project:
0:53

Single video in, 4D human out. 4DAnyone turns a casual monocular video into a 4DGS model. No camera rig, no calibration, no tripod. Project:

Yudong Jin

41,519 views • 13 days ago

Sketch-to-image is now live in Realtime mode on wan video. Draw anything — a stick figure, a wobbly rectangle, three lines you're not even sure about — and watch it render into a photorealistic scene in real time. No export, no waiting, no "generate" button. The canvas just... gets it. Now in Labs on wan video. Go draw something and see what comes back.
0:59

Sketch-to-image is now live in Realtime mode on wan video. Draw anything — a stick figure, a wobbly rectangle, three lines you're not even sure about — and watch it render into a photorealistic scene in real time. No export, no waiting, no "generate" button. The canvas just... gets it. Now in Labs on wan video. Go draw something and see what comes back.

Wan

4,079,241 views • 1 month ago

You don't need a GPU to run text-to-speech anymore. Kyutai released Pocket TTS, a 100 million parameter speech model that runs entirely on CPU. How it performs: > Nearly 6x faster than real time on a MacBook CPU > First audio chunk arrives in about 200 milliseconds > Runs on just 2 CPU cores > Supports voice cloning and 6 languages Setup takes one command: pip install pocket-tts Text-to-speech that used to need a GPU or a paid API now runs locally, on a laptop, in near real time.
0:27

You don't need a GPU to run text-to-speech anymore. Kyutai released Pocket TTS, a 100 million parameter speech model that runs entirely on CPU. How it performs: > Nearly 6x faster than real time on a MacBook CPU > First audio chunk arrives in about 200 milliseconds > Runs on just 2 CPU cores > Supports voice cloning and 6 languages Setup takes one command: pip install pocket-tts Text-to-speech that used to need a GPU or a paid API now runs locally, on a laptop, in near real time.

Simplifying AI

14,328 views • 27 days ago

vLLM v0.24.0 is out! 571 commits from 256 contributors (77 new). 🎉 Highlights: MiniMax-M3 support (FP8/MXFP4 + broad AMD tuning), DeepSeek-V4 keeps maturing (FlashInfer sparse index cache, prefill chunk-planning, now on SM120), Model Runner V2 now handles quantized models by default, a new unified Streaming Parser Engine for tool-calls + reasoning, DiffusionGemma, DeepEP v2 for wide expert parallelism, and a maturing Rust frontend. Thread 👇
0:31

vLLM v0.24.0 is out! 571 commits from 256 contributors (77 new). 🎉 Highlights: MiniMax-M3 support (FP8/MXFP4 + broad AMD tuning), DeepSeek-V4 keeps maturing (FlashInfer sparse index cache, prefill chunk-planning, now on SM120), Model Runner V2 now handles quantized models by default, a new unified Streaming Parser Engine for tool-calls + reasoning, DiffusionGemma, DeepEP v2 for wide expert parallelism, and a maturing Rust frontend. Thread 👇

vLLM

52,567 views • 2 months ago

He just sat on livestream looking at a video on Twitter while logged into it.
3:01

He just sat on livestream looking at a video on Twitter while logged into it.

𝐇𝐎𝐓𝐄𝐏 悪いバナー 👹🇺🇸

11,642 views • 3 months ago

Heading into this final game vs. Marc-Andre Fleury, Sidney Crosby said, "you reflect a bit...he's had an amazing career, and it was fun having an opportunity to play with him for a big chunk of it." Said time goes by so fast, and it feels like yesterday they came into the league.
1:44

Heading into this final game vs. Marc-Andre Fleury, Sidney Crosby said, "you reflect a bit...he's had an amazing career, and it was fun having an opportunity to play with him for a big chunk of it." Said time goes by so fast, and it feels like yesterday they came into the league.

Pens Inside Scoop

110,357 views • 1 year ago