Search “Video”

Krea launches image and video generation bot for Grok

Krea announces the launch of Krea bot for Grok, allowing users to generate images and videos directly within Grok using its models.

PixVerse: Video Input to Motion Transfer Workflow

PixVerse states that precisely prompting motion in AI video is still troublesome; a simpler workflow is: input video → input character → transfer motion, suitable for replicating trending styles without starting from scratch.

AI mass-generates e-commerce videos: a 5-person team produces over a thousand per day, at under 5 yuan each

Yunying Research Society analyzes that AIGC is taking over the mass-production stage of e-commerce recommendation videos; a team of about 5 people can generate over a thousand videos in a few minutes, at under 5 yuan each, while manual shooting costs about 100 to 200 yuan per video. The specific approach is to first break out fixed blocks and variable blocks to form a template, then use low-cost models such as Grok Video and the Seedance entry version to test hooks and directions, and after getting data, switch to Veo, Sora, and the Seedance HD version to produce finished products. The article also mentions that Douyin, Xiaohongshu, and Video Accounts all tightened AIGC content review this year, and that Ocean Engine reported at the end of August that it had shut down more than 50,000 violating accounts, and holds that once materials are no longer scarce, judgment and aesthetics become the new threshold.

Codex automatically edits videos with DaVinci Resolve

The author describes inserting the SD card with the footage into a laptop and then leaving the workstation, letting dot and Codex automatically edit the video with DaVinci Resolve and periodically push progress updates via Slack for review.

German media regulator survey: AI videos significantly weaken public trust in video content

The "Video Trends 2026" survey released by Germany's state media regulators shows that 79% of respondents have encountered AI-generated or heavily modified videos online, most concentrated on TikTok and Instagram; among users who encountered such content, 68% mistook it for real at least once. 81% of respondents said they now more often doubt the authenticity of videos, about 90% demand mandatory labeling of AI content, and 83% support banning certain forms of deepfakes.

fal Releases FLUX 3 Max Video Model, Generating a 10-Second Video Takes 12.45 Seconds

fal, in partnership with bfl_ai, released the post-trained video model FLUX 3 Max, which takes 12.45 seconds to generate a 10-second video, compared with 277.8 seconds for Seedance 2.5; the two share an Elo score of 1,139 on third-party benchmark evaluations. The model is tied for first place in image-to-video quality on an independent third-party benchmark, and can generate 10-second, 720/768p videos in about 12 seconds. fal said FLUX 3 Max will be 75% off for the next three weeks.

Perplexity's Computer generates a popular science video on chip manufacturing

Aravind Srinivas commented that access to experts is no longer the bottleneck to learning new knowledge, and that AI can generate a five-minute video to learn anything. The Perplexity account Computer, which he cited, stated that the video drew on 13 technical references from ASML, Intel, TSMC, Cadence, SkyWater, and IEEE Spectrum, and noted that at 3:59 it explains that "3 nanometers" does not mean what it literally sounds like.

Nectar Uses AI Agents to Parse Brand Mentions in Videos

Nectar has launched an agentic growth stack that can analyze video visuals, voiceovers, and on-screen text across Instagram, TikTok, YouTube, X, and Reddit, breaking each post down into 300+ attributes to capture brand mentions that never appear as text. Nectar estimates that 95% of brand discussions happen outside tagged posts, which is the only content most social tools track.

The author used Grok Bot to generate the MemT-1.0 popular-science animation video purely with code

The author says they gave Grok Bot only one sentence and one avatar image and wrote no storyboard prompts at all, yet in the background it used pure front-end code to compile the newly open-sourced memory large model MemT-1.0 into a one-minute illustrated popular-science animation, with the animation, voiceover, and storyboard all written as code in one go and rendered on the spot. The author believes that video-as-code is easy to extend: to change the avatar you only need to change one image path line, and to switch to an English narration you just re-run the compilation and the timing will automatically realign.

Bengaluru police arrest a man suspected of using AI to generate a fake explosion video of Karnataka's Vidhana Soudha

Bengaluru police arrested 25-year-old Hazarat Ali, who is suspected of posting an AI-generated fake video of an explosion at Vidhana Soudha on Instagram; no explosion actually occurred. Police registered a case under Sections 353(1)(b) and 353(2) of the Bharatiya Nyaya Sanhita, and traced his residence by analyzing digital traces such as IP logs, announcing the arrest on Wednesday, October 21. Police are investigating his motive and whether others were involved, and warned against misusing AI to create threatening content.

Kling 4.0 launches on OnSolo with support for generating 30-second native video in a single pass

Kling 4.0 is now live on OnSolo and can generate 30 seconds of native video in one pass, maintaining motion stability through fast action and large-scale camera movements. This version supports stereo sound and more precise lip sync, a 21:9 widescreen aspect ratio, and up to 15 references and 10 keyframes stacked in a single job to maintain consistency of characters, shots, and story.

Kling Appears at LA Tech Week to Discuss AI Video for Enterprise Applications

At LA Tech Week 2026, Kling joined Oxen AI and Obsidian Studio to discuss how AI video can be integrated into enterprise content production, covering the process from shot creation to final delivery. The post also mentioned how Kling 4.0 supports complex production needs.

MiniMax responds to H3 integration with TPU: looking forward to open-source work on efficient video inference

MiniMax officially responded to SGLang adding MiniMax H3 to the TPU support list of SGLang-TorchTPU, and said it looks forward to more open-source work on efficient video inference. SGLang said that SGLang-TorchTPU was built jointly by SGLang, the Google TPU team, and the Meta PyTorch team, has been used for production inference at Meta, and supports models including Muse Spark Open, GLM-5.3 Flash, Kimi K3, Ling 3.1 Flash, and MiniMax H3.

/video-short skill includes a built-in video review page

Peter Yang said that when giving AI feedback on video edits in a chat, it is hard to pinpoint specific timestamps, so his /video-short skill includes a built-in review page: you can pause on a frame and leave a comment for the AI to fix; the agent applies each piece of feedback before loading the next draft. He described it as being like Google Docs comments for video, and provided a free GitHub link and tutorial.

English translation is being prepared.

English translation is being prepared.

fal claims FLUX 3 Max scored 1,139 Elo in third-party evaluations, ranking in the top tier of image-to-video

fal states that FLUX 3 Max achieved 1,139 Elo in third-party benchmark evaluations, entering the top tier of frontier image-to-video models based on independent human preference assessments. The model was launched through a collaboration between fal Research and bfl_ai; earlier introductions said its image-to-video quality tied for first in independent third-party benchmarks, and it can generate 10-second, 720/768p videos in about 12 seconds. fal Research says it is continuing post-training of the model and will subsequently improve quality and performance.

Tsinghua Team Proposes PPMem: Using Planar Primitives to Build Persistent Memory for AI Video World Models

A team from Tsinghua University has proposed PPMem, which replaces the previous SPMem's volumetric 3D memory with planar primitives to store explored scenes for video world models. Each planar primitive is parameterized by only 7 numbers (center depth, two rectangular extents, and a rotation quaternion), making it orders of magnitude more compact than per-pixel representations. When the camera returns to a previously visited location, the stored planes are projected back into the field of view to guide the generation of new frames, reducing the spatial information that must be retained and processed and preventing familiar objects from changing or disappearing upon revisits.

English translation is being prepared.

MiniMax shares creator's 2D-to-3D video made with H3

The official MiniMax Design account shared creator aicreataro's work, marveling that the creator community always brings surprises. The work first uses Midjourney to generate a KV, then uses MiniMax H3 to generate a 10-second running video, swaps in 7 different materials, and finally builds a 3D space in Blender that matches the 2D footage and connects the two.