Xiang

Software Engineer with background in Structural Engineering. Build to amaze✨.

[Sam @Sam8002642]

Pov : 2 hours before deadline

[⚡AI Search⚡ @aisearchio]

Seedance 2.0. Absolute madness

[Dorksense @Dork_sense]

Seedance 2.0 from China will be the SOTA

This is AI
We are cooked.

• Native multi-shot storytelling from a single prompt (no more stitching scenes)
• Phoneme-level lip-sync in 8+ languages
• 30% faster generation than v1 via RayFlow optimization • 1080p cinematic quality, maintains motion consistency across shots
• Audio-video joint generation = no separate audio pipeline
• Multi-shot coherence = you could generate a full short film scene in one prompt

[歸藏(guizang.ai) @op7418]

字节今天发布了 Seedance 1.5 Pro 视频生成模型

也支持音画同出，比较离谱的是他们支持方言

这一手老陕西人吃面的时候说的陕西话，太牛批了

纯文生视频

[Magnific @magnific]

Introducing Google Veo 3.1

Characters stay consistent, no matter the angle. Cinematic, fast, unreal

Make it UNLIMITED with the link below 👇

Available now on Freepik

[Linus ✦ Ekenstam @LinusEkenstam]

Most people don’t understand how fast AI development is going.

This is Image to 3D world 🤯

[PJ Ace @PJaccetturo]

Veo 3.1 + Nano Banana is insane 🤯

Google’s new models let us make million-dollar looking ads for brands like Wander.

Copy our entire process for making this ad below 👇🧵

[Martin Nebelong @MartinNebelong]

With Veo 3 and Flow out in the world, here's a few examples of videos I've created with Veo 3.

The first video is an example of the incredible voice/audio capabilities. The second one is a test of doing a longer form video (edited in Premiere).

Generated with Veo.

[Bennett Waisbren @BennettWaisbren]

ChatGPT 4o's new image gen is insane. Here's what Severance would look like in 8 famous animation styles 🧵

1/8:
Rankin/Bass – That nostalgic stop-motion look like Rudolph the Red-Nosed Reindeer. Cozy and janky.

[el.cine @EHuanglu]

China's AI is getting crazier..

ByteDance’s new AI model, OmniHuman-1 can generate video with realistic lip sync even if its side view and...

it can control expression and gesture using audio and text..

10 examples:

[Min Choi @minchoi]

Wow.

DeepSeek just dropped Janus-Pro-7B, an open-source multimodal AI that beats DALL-E 3 and Stable Diffusion.

The 🐋 is on fire. 👀

[Xiang @xli9052]

@AlphaBeta_2017 @tj_research lol

[Xiang @xli9052]

Gemini 2.0 real-time multi-modal processing is mind blowing! #Gemini

[AshutoshShrivastava @ai_for_success]

Less than 12 hours of Gemini 2.0 Flash and it has already blown everyone's mind.
Here's how people are using it, here I have collected some of the best post, great use cases and examples 👇

[Xiang @xli9052]

@OpenAI Sora is finally available to the public! And here's the review from MKBHD: https://t.co/44ge6HfUFy
#SORA #aivideo

[Marques Brownlee @MKBHD]

The rumors are true - SORA, OpenAI's AI video generator, is launching for the public today...

I've been using it for about a week now, and have reviewed it: https://t.co/6cDqDQTK5g

THE BELOW VIDEO IS 100% AI GENERATED

I've learned a lot testing this, here are some new learnings. Thread 🧵

[Xiang @xli9052]

@HBCoop_ And open source! 🔥

[Xiang @xli9052]

It’s great to see another promising player in the game! #HunyuanVideo #AIvideo

[Heather Cooper @HBCoop_]

Tencent's Hunyuan Video is the latest open source AI text-to-video model, and the quality is mind-blowing.

5 Examples from my first results:

1. Prompt: Low-angle tracking shot following a sleek BMW motorcycles driving on a busy street in New York City. Street lights reflecting off wet pavement

[Pierrick Chevallier | IA @CharaspowerAI]

Still blown away by the facial expressions #hailuo_ai can generate 🤯🎥. The level of detail and realism is just insane 🔥

High-angle tight shot, camera hovers above a man kneeling in the rain, his face upturned toward the sky. Tears mix with the rain on his cheeks, and his open mouth quivers in a silent scream

[Aakash @aakashsastry]

Today @HotshotSupport we're launching Reference Videos on Hotshot - a new way to control your Dancing, Acting, Reactions, and more.

Try it now at https://t.co/jEnopEKlRY

[Xiang @xli9052]

Finally here we go! Official “controlnet”, “ipAdaptor” and in/out paint models for FLUX dev and pro. #flux #aiimage

[Black Forest Labs @bfl_ai]

Today, we are excited to release FLUX.1 Tools, a suite of models designed to add control and steerability to our base text-to-image model FLUX.1, enabling the modification and re-creation of real and generated images. Learn more in our blogpost: https://t.co/J5Bc8fVGEc