Hexiang (Frank) Hu

@xAI: making @grok imagine
| Previously @GoogleDeepMind: Gemini & Imagen
| Seattle ← Los angeles ← Hangzhou ← Wenzhou

[Hexiang (Frank) Hu @hexiang]

The goal is to democratize film making

[Grok @grok]

Grok @Imagine 1.5 Preview is here

Try it today in the API: https://t.co/x4Yt13xRu7

[Hexiang (Frank) Hu @hexiang]

@Dimdv99 In progress

[Hexiang (Frank) Hu @hexiang]

The goal is to democratize film making

[Grok @grok]

Grok @Imagine 1.5 Preview is here

Try it today in the API: https://t.co/x4Yt13xRu7

[Hexiang (Frank) Hu @hexiang]

@amytam01 @SpaceX @xai Welcome

[Hexiang (Frank) Hu @hexiang]

A recent update from the team, enjoy!

[Arena.ai @arena]

Grok-Imagine-Video-1.5-Preview (720p) has landed #1 in the Image-to-Video Arena!

This is a massive +52 pt improvement over Grok-Imagine-Video (720p), surpassing the best video models Seedance-2.0 and HappyHorse.

Congrats to @xAI and @elonmusk on this big achievement!

[Hexiang (Frank) Hu @hexiang]

Imagine would improve a lot on real-world usefulness!

[Grok @grok]

Prototyping game assets directly with Grok @imagine

[Kaylee George @krgeorge]

Inline image & video on alpha release btw (next stable)

[Cursor @cursor_ai]

Together with SpaceXAI, we’re training a significantly larger model from scratch, using 10x more total compute.

With Colossus 2’s million H100-equivalents and our combined data and training techniques, we expect this to be a major leap in model capability.

[Yun-Ta Tsai @yunta_tsai]

Grok Imagine is the new Leica.

Left: Sony Alpha VII
Right: Grok Imagine

[Yun-Ta Tsai @yunta_tsai]

Wow, Grok Imagine knows how to cast the shadow through the glass. It even casts the shadow on the door hinge.

Picture on the left is original (real), right is Imagine.

[Hexiang (Frank) Hu @hexiang]

@mycharmspace Best luck Tianyi!

[Hexiang (Frank) Hu @hexiang]

@md_kasif_uddin It is must easier to name the worst

[Hexiang (Frank) Hu @hexiang]

Congrats on the release @linluqiu, awesome work!

[Linlu Qiu @linluqiu]

Language is discrete. Language models don’t have to be.

🧚Introducing ELF🧚‍♀️: Embedded Language Flows—a class of diffusion models in continuous embedding space based on continuous-time Flow Matching 🧵

[Hexiang (Frank) Hu @hexiang]

🧐🧐

[The Humanoid Hub @TheHumanoidHub]

Unitree CEO Wang Xingxing just unveiled a real-life mecha.

Marketed as the world’s first mass-produced manned robot, this machine can transform into a quadrupedal civilian vehicle. The unit weighs roughly 500 kg (1,100 lb), including the pilot.

[Hexiang (Frank) Hu @hexiang]

@archanfel_anoth Best luck 🐐🫡

[fal @fal]

🚨 Grok Imagine Quality Mode API is now available on fal!

🖼️ High-quality image generation from text prompts
✏️ Edit existing images with natural instructions
🔍 Analyze visuals and reason about image content

[xAI @xai]

Image Generation Quality Mode is now available on the xAI API.

This model has already powered the generation of over 300 million images on Grok.

It brings higher realism, stronger text rendering, and better creative control for business professionals.

https://t.co/se5D7T5RJW

[Hexiang (Frank) Hu @hexiang]

@IamEmily2050 @keepgoingAnnie It can not be the end of the year 🤣

[Hexiang (Frank) Hu @hexiang]

We are making an upgrade for our image gen API 🚀🚀!!

Also a blogpost is here to show how to best utilize its ability: https://t.co/2tl0l31Jba