Jie Huang (@jefffhj) header
Jie Huang (@jefffhj) profile picture

Building intelligence @xAI. Grok-2🍍, 3🍫, 4🫐, Video Gen🪄

Joined July 2017
Photo by Jie Huang (@jefffhj)Photo by Jie Huang (@jefffhj)Photo by Jie Huang (@jefffhj)Photo by Jie Huang (@jefffhj)Photo by Jie Huang (@jefffhj)Photo by Jie Huang (@jefffhj)Photo by Jie Huang (@jefffhj)Photo by Jie Huang (@jefffhj)Photo by Jie Huang (@jefffhj)Photo by Jie Huang (@jefffhj)
This is the second #1 I achieved at @xAI! (the first one was Grok 3 with my best post-training friends ;). After shipping Grok 4, I jumped into something completely new to me — video generation with @imhaotian and @zeliu_. In just 6 months, the team built everything from 0 to 1 to #1. How did we make this happen? For me, it's mostly about the best intuition and execution — and having the best friends to work with. My best take on both achievements is that we should respect users and focus on things that can scale. This is just the start. Stay tuned for what we’ll achieve next.
xAI's Grok Imagine takes the #1 spot in both Text to Video and Image to Video in the Artificial Analysis Video Arena, surpassing Runway Gen-4.5, Kling 2.5 Turbo, and Veo 3.1! Grok Imagine is the latest video model from @xAI, and joins an increasing roster of models such as Google's Veo and OpenAI's Sora supporting native audio generation alongside video output. The model is available via xAI's new Grok Imagine API only. It is priced at $4.20 per minute including audio, matching Kling 2.5 Turbo (without audio) and cheaper than Veo 3.1 Preview ($12/min) and Sora 2 Pro ($30/min). Congratulations @elonmusk and the xAI team. See below for comparisons between Grok Imagine and other leading models in our Artificial Analysis Video Arena
16
10
3
239
27,156
Try Grok Imagine Video 1.5 Lite — faster, cheaper, and with good quality.
Try Imagine Lite. Same quality at around 20% of the cost
4
1
28
1,662
Jie Huang retweeted
Try Imagine Lite. Same quality at around 20% of the cost
Video 1.5 Lite is live in the Grok Imagine API. The new workhorse for text-to-video and image-to-video. $0.02/sec at 480p $0.03/sec at 720p $0.14/sec at 1080p
1
1
1
19
2,888
Very impressive!
Congrats to the launch 🚀 and excited to see SGLang @sgl_project and Miles @radixark helped Periodic Lab’s infrastructure and research!
1
5
2,024
Jie Huang retweeted
Grok Imagine Video 1.5 Agent from @spacexai has landed in the Text-to-Video Arena at #5 with 1491 pts! This release performs on par with Wan-3.0 and FLUX 3 Video (each with 1494 pts), just 3 pts below. Grok Imagine Video 1.5 Agent out performs both Dreamina Seedance-2.5 and 2.0, and MiniMax-H3. @spacexai is now in the top 5 labs in the Text-to-Video Arena. Congrats to the team!
Grok Imagine Video 1.5 agent is now available. Powered by our newest Image 2.0 model, it delivers higher quality, better storytelling from a smarter agent and excels at connecting multiple shots together with greater continuity.
20
21
7
306
40,531
Thanks @zhang_yunzhi for leading agent intelligence. Stay tuned for what's coming next :)
Very excited to ship Grok Imagine Video 1.5 agent! @spacexai is constantly pushing the frontier of intelligence and usefulness. More to come.
2
1
1
62
10,515
Jie Huang retweeted
Very excited to ship Grok Imagine Video 1.5 agent! @spacexai is constantly pushing the frontier of intelligence and usefulness. More to come.
Grok Imagine Video 1.5 agent is now available. Powered by our newest Image 2.0 model, it delivers higher quality, better storytelling from a smarter agent and excels at connecting multiple shots together with greater continuity.
13
11
3
229
25,391
Jie Huang retweeted
Grok Imagine Video 1.5 agent is now available. Powered by our newest Image 2.0 model, it delivers higher quality, better storytelling from a smarter agent and excels at connecting multiple shots together with greater continuity.
347
482
112
3,120
1,212,507
Jie Huang retweeted
Grok 4.6 is more agentic, better at hard problems, and stronger on knowledge work, while staying just as fast and inexpensive. Please try it on your toughest tasks and share critical feedback: we plan to ship rapid improvements!
Introducing Grok 4.6. It delivers frontier intelligence and is a significant improvement over Grok 4.5 at the same price.
9
25
1
239
15,522
🚀
Introducing Grok 4.6. It delivers frontier intelligence and is a significant improvement over Grok 4.5 at the same price.
1
22
1,955
Try Imagine Image 2.0!
Announcing Imagine Image 2.0, our next generation image model with precision editing, crisp text rendering, improved factuality, and real world usefulness. Image 2.0 helps you make images for real work. x.ai/news/grok-imagine-image…
7
38
2,754
Jie Huang retweeted
Exciting news: Grok Imagine Image 2.0 (Low) by @spacexai has landed in the Text-to-Image Arena at #2 (1320 pts)! This release is not available via API, only in their app. Grok Imagine Image 2.0 (Low) is a significant improvement from Grok Imagine Image Quality (#14 -> #2). By category, it’s #3 in Product, Branding & Commercial Design, and #2 across: - Art - Portraits - Text Rendering - Cartoon, Anime & Fantasy - Photorealistic & Cinematic Imagery In the Image Edit Arena, Grok Imagine Image 2.0 (Low) also debuted at #2 (1439 pts)! See post below. Congrats to the @spacexai team on this release!
Announcing Imagine Image 2.0, our next generation image model with precision editing, crisp text rendering, improved factuality, and real world usefulness. Image 2.0 helps you make images for real work. x.ai/news/grok-imagine-image…
60
111
40
1,200
280,712
🐐
Grok Imagine is evolving rapidly. Today, a great image model needs to go beyond aesthetics, it must also deliver factual accuracy and real-world usefulness. Achieving this requires reasoning, planning, factual grounding, and self-reflection. We’re planning to release several tiers of new image models, each offering different levels of capability and intelligence. Today, we’re introducing the base version, with more to come over next days. try it out in grok.com/imagine
24
2,133
Jie Huang retweeted
Announcing Imagine Image 2.0, our next generation image model with precision editing, crisp text rendering, improved factuality, and real world usefulness. Image 2.0 helps you make images for real work. x.ai/news/grok-imagine-image…
6,123
4,426
1,293
38,755
52,128,396
Try Grok Imagine Video 1.5 with omni-reference
Imagine Video 1.5 launched last month as our best video model yet, with more lifelike motion and sound. Today it goes even further with text-to-video support, image and voice references, and native 1080p. grok.com/imagine
11
2,183
Jie Huang retweeted
For my first post, I’m sharing a letter @NVIDIA signed on why open models matter. AI will transform every industry, power every company, and be built by every country. Open models strengthen safety and cybersecurity, accelerate innovation and diffusion, and enable sovereignty. The world needs both frontier closed models and frontier open models. images.nvidia.com/pdf/Open-W…
16,548
34,449
12,082
193,760
69,458,739
Jie Huang retweeted
Splashdown confirmed! Congratulations to the entire SpaceX team on the 13th flight test of Starship!
1,257
4,435
804
34,240
2,391,466
It's a good model.
Announcing Grok 4.5, our first model trained specifically for coding and agents. It was trained with Cursor and offers frontier intelligence at leading speeds and cost efficiency. x.ai/news/grok-4-5
2
65
2,576
Announcing Grok 4.5, our first model trained specifically for coding and agents. It was trained with Cursor and offers frontier intelligence at leading speeds and cost efficiency. x.ai/news/grok-4-5
1,485
3,447
1,168
26,449
69,314,800
Jie Huang retweeted
Our latest image to video model now GA 🔥 Also meet Imagine Video 1.5-fast and many new features for creative workflows on grok.com/imagine Stunning 720p videos generated in just 25 seconds! 🎥
Grok Imagine Video 1.5 is here Our new image-to-video model with sharper realism, better physics and faster generations 🧵 grok.com/imagine
22
15
160
14,391
Grok Imagine Video 1.5 is here Our new image-to-video model with sharper realism, better physics and faster generations 🧵 grok.com/imagine
1,044
1,278
310
8,150
68,395,057