Skip to content
Date and source

Dates follow the news timeline in Beijing time and include both endpoints.

Nov 8Sun
  1. Luma News19

    HeyGen's Limitations: What It Cannot Do

    HeyGen cannot create cinematic scenes, product shots, or environmental content; it only supports preset gestures, with no custom animation or full-body motion control. During peak rendering times, delays can reach up to 60 minutes. Lip-sync accuracy decreases with tonal languages and regional accents, and it does not provide HDR or EXR professional color-grading output. Luma's Ray 3.2 supports 16-bit EXR export, Layers supports element-level editing, and Luma Agents can maintain project context.

Nov 7Sat
  1. Higgsfield 官方博客60

    Cinema Studio 4.0: The Complete Guide to Cinematic AI Video Production

    Higgsfield releases Cinema Studio 4.0: The Complete Guide to Cinematic AI Video Production, explaining through aperture, lighting, and camera movement settings how to make the visuals serve the scene's objective. The guide uses same-scene comparison tests to show the differences between f/1.4 and f/11, Contre jour and Soft cross, Handheld and Static shot, and provides reusable prompt structures and setting combinations.

    Why it matters: Using same-scene comparison tests, it demonstrates the impact of aperture, lighting, and camera movement on the mood of the visuals, and provides reusable prompt structures and setting combinations.

Nov 6Fri
  1. Reallusion38

    AI Studio Storytellers: Empowering 3D Character Artists with Storytelling Capabilities

    Peter Alexander (Mythcons) leverages Reallusion AI Studio to combine 3D scene construction with AI video generation, transitioning from character asset creator to director. He uses ZBrush, Blender, and iClone for character modeling, scene layout, and frame anchoring, making AI Studio an animation assistant. This hybrid workflow solves AI consistency and proportion issues, enabling independent artists to complete full animated scenes.

  2. PixVerse 中文官方博客69

    PixVerse's hands-on comparison of the best Serverless AI video inference service providers in 2026

    PixVerse releases a comparison of the best Serverless AI video inference service providers in 2026, covering five providers: fal, PixVerse, Baseten, Runpod, and Modal, and uses the same product image to test three models: PixVerse V6, H3 Max, and Wan 2.2.

    Why it matters: PixVerse uses the same product image to test three models, providing a comparison of per-generation cost, speed, and quality, and includes a reusable testing process for service providers.

  3. Kapwing 创作指南64

    7 video editing tasks that can be automated with ChatGPT or Claude

    Kapwing tested automating seven types of video editing tasks with ChatGPT or Claude, covering editing planning, subtitle generation, translation dubbing, and B-roll production. In the test, Claude Code with FFmpeg built and rendered a 70.7-second YouTube Short, while Kapwing MCP completed AI dubbing in five languages. The article notes that standard chat interfaces cannot directly edit videos; actual editing requires a coding agent or an editor connected via MCP.

    Why it matters: The article provides seven types of automatable video editing tasks along with corresponding prompts and tool choices, allowing creators to build their own AI editing workflows based on them.

  4. Kapwing 创作指南65

    How to Use ChatGPT to Edit Videos (2026 Guide)

    Kapwing MCP connects ChatGPT to a video editing workspace, allowing users to complete editing tasks such as uploading, subtitling, translating, dubbing, and generating B-roll using natural language. In testing, a single combined request generated six exported versions (the original English version and five dubbed versions).

    Why it matters: Kapwing MCP allows ChatGPT to directly operate a video editing workspace. The tutorial demonstrates the complete natural language editing workflow from uploading, subtitling, translating, dubbing, to generating B-roll, and provides reusable prompt examples and cost references.

  5. Higgsfield 官方博客45

    How Cinema Studio 4.0 maintains consistency of characters and locations in AI shots

    Cinema Studio 4.0 uses the Elements feature to save characters, locations, and props as references, and adds them to each prompt with @ tags to maintain consistency across shots. A single generation accepts up to 50 references; in testing, a 20-second generation maintained the same cave and characters across 5 shots. This feature is suitable for characters and locations reused across projects or scenes.

  6. World Labs · 官方研究与公告60

    World Labs releases Atlas real-time research preview, Frameboy supports 15fps streaming exploration

    World Labs releases Atlas real-time research preview, demonstrating via Frameboy streaming frame generation from cloud GPU to browser at up to 15fps. This version highly optimizes Atlas while maintaining novel view synthesis capability; it is also integrated into Viewfinder, allowing users to build scenes frame by frame and then explore in real time.

    Why it matters: Real-time streaming generation reduces exploration cost to near zero, enabling creators to assess the potential of interactive world models to replace existing workflows.

  7. Google AI Blog82

    Gemini Omni 1.2 Flash: Faster, more cost-effective, supports 15-second clips.

    Google releases Gemini Omni 1.2 Flash, which natively supports 15-second video generation; compared to 1.1, the price drops 30% to $0.07/second (720p). The new model supports multi-turn conversational editing, enhanced reference following, and can be upscaled to 1080p or 4K after 360p/720p drafts. Gemini Omni is now generally available in Gemini API and Gemini for business.

    Why it matters: The new model increases native video length to 15 seconds and cuts prices by 30%, making it suitable for evaluating batch video production and ad creative production costs.

Nov 5Thu
  1. 可灵官方博客78

    TechHalla reviews Kling 4.0 and 4.0 Flash: control and speed

    Creator TechHalla evaluates Kling 4.0 and 4.0 Flash in a real project, saying 4.0 is for control and image quality, while 4.0 Flash is for speed and batch. He uses 3D animation as motion reference and images as identity reference, generating weapon animation in three steps with timing preserved; multi-shot hard cuts allow 16 generations to complete an entire music video, and 19 generated characters all pass consistency on the first try.

    Why it matters: Creator evaluates Kling 4.0 and 4.0 Flash in a real project, demonstrating transferable workflows such as multi-shot hard cuts, 3D motion reference, and character consistency.

  2. 广东省广电局 · 通知公告53

    Guangdong Provincial Radio and Television Bureau announces the first batch of 39 proposed support projects for micro-dramas and online animation films in 2026.

    Guangdong Provincial Radio and Television Bureau announces the first batch of 39 proposed support projects for micro-dramas and online animation films in 2026, with the public notice period ending on November 11. According to the notice Yue Guang Dian Wang [2026] No. 13, these projects were selected through a special review process. During the public notice period, any objections may be submitted in writing to the Provincial Bureau's Network Audiovisual Program Management Division.

  3. Luma News25

    MiniMax H3 alternatives: 7 omni-modal models that require no self-hosting

    MiniMax H3 requires a self-hosted GPU cluster; 7 omni-modal platforms can replace it without operations overhead. Luma Ray 3.2 leads, supporting multi-keyframe sequences, 1080p watermark-free export, and 4x faster generation, with direct export to Premiere, DaVinci, and Final Cut; Google Veo 3.1 can generate synchronized speech and sound effects in one pass, and ByteDance Seedance 2.5 targets multi-asset brand projects.

Nov 4Wed
  1. Kapwing 创作指南71

    Kapwing Guide: When to Label Videos with AI Labels

    Kapwing has released a creator guide with a guiding principle: AI-generated content that viewers might mistake for real should be labeled with an AI label. The guide summarizes the differing rules on platforms such as TikTok, Instagram, Facebook, and YouTube, and cites research showing that AI labels can affect engagement, recommending that creators add their own wording to the disclosure.

    Why it matters: It provides a general principle to apply: AI-generated content that could be mistaken for real should be disclosed, and it compares platform rule differences to help creators decide when to label.

  2. 可灵官方博客63

    Ash Koosha: Kling 4.0's long takes give emotions time to unfold

    Ash Koosha shares his experience using Kling 4.0, saying that long takes give emotions time to unfold, and targeted editing can fix individual details. He estimates that discarded footage will be reduced by at least 50%, and Kling 4.0 supports single generations of up to 30 seconds. He believes that more stable camera movement and more nuanced expressions are more important for filmmaking.

    Why it matters: Creators tested Kling 4.0's long take and targeted editing capabilities, estimating a 50% reduction in discarded footage, providing a reference for cinematic long take production and detail correction.

Nov 3Tue
  1. Synthesia · 官方博客60

    Synthesia launches a plugin in ChatGPT: create videos without leaving the chat

    Synthesia launches a plugin in ChatGPT, allowing users to generate videos directly in the chat, covering scenarios such as product launches, training, and marketing. By default, it uses the Syren format, which includes motion graphics, digital humans, voiceover, music, and B-roll; entering 'make it in Studio' generates the video precisely according to the script and opens it in the editor for adjustments. The plugin is available to all Synthesia plans and can integrate with connected plugins to automatically generate videos based on Slack messages or schedules.

    Why it matters: Synthesia embeds video generation into ChatGPT conversations, supporting automation triggered by Slack, schedules, etc., and enables evaluation of whether to incorporate it into existing content production workflows.

  2. MiniMax · 音视频模型66

    MiniMax releases H3-Image, an open-source image generation and editing model

    MiniMax releases open-source image generation and editing model MiniMax-H3-Image, unifying text-to-image and image editing in a single checkpoint. The model is a 33.1B-parameter DiT, distilled to 6-step sampling, directly generating approximately 16-megapixel images (4096×4096 to 6272×2688) without super-resolution.

    Why it matters: The new model unifies text generation and image editing in a single checkpoint, directly outputs 16-megapixel images with only 6-step sampling, allowing evaluation of its suitability for existing image generation workflows.

Nov 2Mon
Oct 31Sat
  1. Higgsfield 官方博客76

    Higgsfield's real-world test of 2026 AI video costs: a 10-second finished video ranges from $1.35 to $7.85

    Higgsfield's test shows that a single generation of a 10-second AI video costs about $7.85 (160.125 credits) on Seedance 2.5 at 1080p, and about $1.35 on Kling 4.0 at 720p. The actual cost must also account for retakes, image preparation, and post-production.

    Why it matters: Provides a method and budget formula for calculating the true cost of AI videos under per-generation pricing, allowing creators to plan retake counts and monthly output accordingly.

  2. Higgsfield 官方博客36

    Higgsfield launches a 7-day unlimited generation promotion: models, plans, and restrictions.

    Higgsfield launches a 7-day unlimited generation promotion. Pro, Plus, Max, and Ultra subscribers can generate unlimited specified image and video models for a limited time. The promotion covers Seedream 5.0 Pro, GPT Image 2.5, Nano Banana Pro, and Seedance 2.5. Available models and resolutions vary by plan and monthly/annual billing cycle.

  3. Podcastle37

    How to convert AI-generated 3D models into social media videos using Meshy and Async

    The combination of Meshy and Async can convert AI-generated 3D models into social media videos. The process is: generate a 3D model in Meshy and render it as a 2D image, then use Async to animate the image, add music and subtitles, and output to Instagram, TikTok, or YouTube Shorts. For short videos, Meshy and Async are sufficient; when fine animation is needed, Blender or DeepMotion can still be considered.

Oct 30Fri
  1. ElevenLabs 产品与研究60

    ElevenLabs launches SFX v3 sound effects model, now available on ElevenCreative and ElevenAPI

    ElevenLabs launches SFX v3 sound effects model, now available on ElevenCreative and ElevenAPI. SFX v3 handles prompts describing multiple sounds more reliably, produces more believable music and vocal generation, and delivers cleaner, more natural audio quality than SFX v2. Creators can use it in Flows, the Sound Effects tab, Studio, and Music Editor, while developers can integrate it via ElevenAPI.

    Why it matters: SFX v3 improves multi-sound prompt following, music and vocal generation, and audio quality, allowing creators to evaluate whether to replace their existing sound effects generation workflow.

  2. 可灵官方博客29

    What is an Animatic and how to create one with Kling AI?

    An animatic is a moving preview created by arranging storyboard frames along a timeline, used to assess pacing and editing points before production. Kling AI's Kling 4.0 video generation and editing model supports video generation from 3–30 seconds, and with Omni Reference it can combine up to 15 reference materials (10 images, 5 clips totaling 30 seconds of video, 7 subject references), helping creators quickly generate dynamic drafts from storyboards and adjust action, camera movement, and scene pacing.

Oct 29Thu
  1. 可灵官方博客71

    Ami discusses how Kling 4.0 brings subtle emotions to life.

    Kling invited creator Ami to use Kling 4.0 to produce the short film "Tea When You Have Time". She shared her experience using new capabilities such as 30-second generation, Omni Reference, and multi-keyframe. Kling 4.0 supports single-run 30-second generation, 21:9 aspect ratio, up to 10 keyframes, consistent 12-character assets, good lip-sync for Cantonese dialogue, and supports multiple dialects and English accents.

    Why it matters: Creator hands-on testing shows Kling 4.0 performs stably in 30-second long takes, 12-character consistency, and Cantonese lip-sync, providing a basis for evaluating an upgrade.

  2. Reallusion60

    Character Creator 5.2 Released: From AI-Generated Meshes to Animatable 3D Characters

    Reallusion releases a free update for Character Creator 5.2, integrating AI 3D model generation, character wrapping, auto-rigging, wearable clothing, and real-time animation into a unified workflow. The new version integrates Tripo and Hunyuan generation models, supports converting almost any humanoid mesh into CC native topology, and provides an AI credit system to manage 3D generation, motion capture, and AI image/video/audio services.

    Why it matters: CC 5.2 integrates AI generation, wrapping, rigging, clothing, and animation into a single pipeline, allowing creators to evaluate whether to replace multi-tool workflows.

  3. 国家广电总局 · 公告公示44

    Notice from the General Office of the National Radio and Television Administration on the distribution licensing status of domestic micro-dramas and online animation films nationwide from July to September 2026

    The General Office of the National Radio and Television Administration released the distribution licensing status of domestic micro-dramas and online animation films nationwide from July to September 2026, issuing a total of 338 titles and 7,100 episodes. Among them, micro-dramas accounted for 158 titles and 5,369 episodes, and online animation films accounted for 180 titles and 1,731 episodes. Micro-dramas were predominantly urban themes (38 titles) and revolutionary themes (20 titles), while online animation films were mainly legendary themes (76 titles) and youth themes (54 titles).

  4. HeyGen Updates67

    HeyGen Voice Launches Speech-to-Speech Feature

    HeyGen Voice launches a speech-to-speech (Speech-to-Speech) feature. Users can record a line with their phone or computer and get studio-quality voiceover that preserves emotion, rhythm, stress, and pronunciation, changing only the voice itself. The feature is now available on HeyGen and the API, supporting clean narration from casual recordings and offering a richer emotional range than text prompts, including whispers, laughter, and accents.

    Why it matters: HeyGen's speech-to-speech turns phone recordings directly into studio-quality voiceover, reducing repeated TTS generation, and creators can use it to evaluate the new voiceover workflow.

Oct 28Wed
  1. Kapwing 创作指南66

    Kapwing Tutorial: How to Make a Tiny Desk AI Trend Video (with Prompts)

    Kapwing releases a Tiny Desk AI trend production tutorial, using Seedance 2.5 to generate four video clips and compositing them in a Kapwing template. The tutorial includes character reference images, motion references, and group/solo prompts, reproducing the shots and movements of Charlie Wilson's 2023 NPR Tiny Desk performance of "Outstanding".

    Why it matters: The tutorial provides a complete workflow with character reference images, motion references, and group/solo prompts, addressing character consistency and instrument distortion issues in AI generation.

Oct 27Tue
  1. Higgsfield 官方博客72

    Kling 4.0 and 3.0 prompt guide: 10 usable prompts and settings.

    Higgsfield released a prompt guide for Kling 4.0 and 3.0, comparing their applicable scenarios and providing 10 copyable prompts. Kling 4.0 supports up to 30 seconds, 10 keyframes, automatic multi-shot, and video editing; Kling 3.0 supports up to 15 seconds, 5 shots, Elements, and native audio. The guide recommends starting prompts with subject, action, and camera, then adding time, reference, lighting, audio, and lock for complex scenes.

    Why it matters: This guide compares the applicable scenarios of Kling 4.0 and 3.0, and provides 10 copy-ready prompts to help creators choose the right model for their tasks.

  2. 可灵官方博客34

    Kling 4.0: 8 Common Camera Angles and Their Uses

    Kling 4.0 supports controlling camera angle and motion through prompts and visual references; the Omni Reference feature can load up to 15 reference assets. The model can generate videos from 3 to 30 seconds and stably handle complex camera movements such as tracking, orbiting, panning, pushing in, and pulling out. The AI video editor supports targeted adjustments across up to 5 clips with a total duration of 30 seconds.

  3. 可灵官方博客15

    ChatGPT review: Pros, cons, and real-world performance.

    ChatGPT excels at drafting, research, coding, file analysis, and image creation, but complex visual projects may require specialized tools. The review compared ChatGPT Images 2.5's conversational generation and editing with Kling IMAGE 3.0 Omni (supports 10 reference images, Image Series, native 2K/4K output) and Kling 4.0 (Omni Reference supports 15 reference assets, 3-30 second video generation).