Skip to content
Date and source

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

Nov 3Tue
  1. AI Fruit · 中文图像视频制作指南68

    HyperFrames GitHub Repository Guide: Installation Tutorial and Comparison with Remotion

    HyperFrames is an open-source HTML-to-video framework by HeyGen. The author tested its installation and rendering workflow and compared it with Remotion. On a Mac, a 10-second 1920×1080 video rendered in under 20 seconds, producing deterministic MP4 output suitable for visual regression in CI. The framework is Apache-2.0 licensed with no per-render fees, making it suitable for agentic creation; Remotion has a more mature ecosystem and is better suited for React teams.

    Why it matters: The author tested HyperFrames' complete rendering workflow and compared it with Remotion in terms of licensing, ecosystem, and agent suitability, helping creators choose the right tool.

  2. AITuber · AI短视频与提示词36

    AI Slop vs. Lore: How Faceless Channels Maintain Originality

    Sprout Social named "lore" as the 2026 social media word of the year, with over 10 million mentions; "slop" came second with 8.8 million mentions, reflecting resistance to mass-produced AI content. For faceless AI channels, audiences punish repetitive faceless filler content and reward series with characters and story worlds. YouTube policy also points in this direction: series with each episode having an independent focus are viable, but templated mass production is not.

  3. Reddit comfyui · 创作与工作流分享69

    ComfyUI Tutorial: PIKA H3 + MiniMax H3 Singularity V2 + NVIDIA Regenerative 2K Super-Resolution KIJAI LORA (6GB VRAM)

    The author shared a ComfyUI workflow that uses MiniMax H3 Singularity V2 and Pika H3 LoRA to generate a 0.6MP video in 6 steps, then enhances it to Full HD via regenerative 2K super-resolution. Total generation plus super-resolution time is about 1 hour, final resolution 1920×1080, runs on 6GB VRAM. The author provided the required LoRA, custom nodes, and workflow links.

    Why it matters: This workflow demonstrates a path of low-resolution generation combined with regenerative super-resolution enhancement, allowing Full HD video to be produced with only 6GB of VRAM, and the method can be transferred to other models.

  4. Leeveo AI Tutorials · 创作教程与项目分享61

    How to use AI to turn a selfie into a movie character: A seven-view character reference sheet tutorial

    Lee Veo released a tutorial showing how to use AI to transform an ordinary selfie into a hyper-realistic movie character and generate a seven-view character reference sheet. The tutorial covers character design briefs, main reference image generation, seven-view reference sheet creation, and consistency improvement techniques, and recommends using tools such as WAN, Seedance, or MiniMax to animate the reference sheet.

  5. note · seisei_ai_1st · 图像视频创作分享15

    Steps for Grok to create adult-oriented spicy AI videos and copyable system prompts

    Grok chat can create adult-oriented "spicy" AI videos; the tutorial provides two production workflows and copyable system prompts. The cheapest Standard plan for the video generation service used is $12/month, with 40% off for annual payment. The system prompts can guide Grok from listening to requirements to generation.

  6. 雷峰网71

    Hands-on with Kling 4.0: How can we get closer to the ideal effect when making AI dance?

    Leiphone tested Kling 4.0 and found that complex dynamic generation is still difficult to stabilize; in the water-sleeve dance, seven or eight rounds failed to stably achieve the sleeve tail hitting the drum. The first version of 'Singin' in the Rain' had a good degree of completion. Video editing to change into a suit succeeded, but the credit consumption was comparable to regeneration. The 30-second life vlog had detail issues such as shot transitions.

    Why it matters: The test shows that Kling 4.0's complex dynamic generation is still difficult to stabilize. Although video editing succeeded, the cost was comparable to regeneration, providing a reference for creators to evaluate practical usability.

  7. note · tsugiito_design · 图像视频创作分享41

    Male knight with cape fluttering in the wind | Creating tags and JSON for black hair, silver armor, blue eyes using ComfyUI

    ComfyUI and Animagine XL 4.0 Opt generate an image of a male knight with black hair, blue eyes, silver armor, and a red cape fluttering in the wind, resolution 1216×832. No reference image, additional models, or post-processing were used. Summarized the prompt, generation settings, and node configuration JSON. Includes comparison outputs of red/blue capes, wind effects, and expressions under the same seed. Model from Cagliostro Labs, 6 files in total.

  8. note · dreammii · 图像视频创作分享37

    AI animation scenes not continuous? How to create using the '3-panel storyboard' method [for beginners]

    In AI animation production, using the '3-panel storyboard' (beginning, change, afterglow) to plan scenes can solve the problem of discontinuous visuals. First, determine a one-line theme, then divide it into 3 panels according to 'beginning, change, afterglow', use ChatGPT to organize the prompts, generate each panel, and check the transitions. This method is suitable for beginners; no hand-drawing is required, and text descriptions can serve as the basis for the storyboard.

  9. lilting channel (English) · 创作分享73

    Anima-RGBA on M1 Max: direct transparent PNG output and alpha limitations

    In actual testing on M1 Max, Anima-RGBA can directly output transparent PNGs and cleanly remove backgrounds, but semi-transparent elements such as glass, steam, wings, and ghosts remain opaque. Background removal works well, but intermediate alpha values are almost nonexistent; raising the trigger word weight to 1.3 also fails to produce semi-transparent effects. With Turbo LoRA added, 8-step generation time is reduced to 41 seconds, making it suitable for everyday cutout workflows.

    Why it matters: Actual testing shows Anima-RGBA can directly output transparent PNGs and cleanly remove backgrounds, but semi-transparent elements such as glass and steam remain opaque, so creators need to make trade-offs based on their use case.

Nov 2Mon
  1. note · kongo_jun · 图像视频创作分享51

    Trying HairDeck in ComfyUI to generate bound hair

    The author tried HairDeck in ComfyUI, generating bound-state hair from a character image, with generation taking 38 seconds and using 8GB of VRAM. The author provided installation commands and a workflow, and mentioned the GLB version. The author said the tool is very helpful for people who don't know how to model, and looks forward to future updates that better reproduce the input image.

  2. 钛媒体78

    Kling 4.0 Hands-on Test: Zero Real People on Camera, Handcrafted Exhibition Video, How Far Is AI from 'One-Stop Video Production'?

    Singularity Research Lab tested Kling 4.0 to produce an exhibition video, handing all real-person appearances to AI. The results were mixed: some shots had quite natural character blending, camera movement, and composition, but other segments had issues such as subject replacement, continuous takes turning into hard cuts, and audio becoming garbled after modifications. The test combined real exhibition footage with AI-generated characters, split into 7 sets of shots, and prepared about 3,000 characters of prompts. The author believes Kling 4.0 has raised the ceiling for generating good shots, but creators still need to backstop key steps to reliably deliver videos that meet requirements.

    Why it matters: The test shows Kling 4.0 can produce professionally usable footage, but reliably executing complex requirements still requires human backup. Creators can use this to evaluate whether to replace their existing video production workflow.

  3. CoAnimator · 可编辑视频与动画工作流41

    Comparison of Gemini CLI and Claude Code in video production

    Claude Code has documented integration in CoAnimator, while Gemini CLI is more suitable for teams that already have enterprise or paid API access. On June 18, 2026, Google migrated free individual users and Google AI Pro/Ultra subscribers from Gemini CLI to Antigravity CLI.

  4. Reddit comfyui · 创作与工作流分享66

    Pika H3 Flex: 380% faster than 30-step H3, differences under the same seed

    Pika H3 Flex's 6-step LCM setting is about 4.8 times faster than the 30-step base sampling; under the same seed, rendering time drops from 27.4 minutes to 5.7 minutes. This LoRA is a distilled model, and the template's 6-step LCM schedule is designed for it. The test was conducted on ComfyUI 0.43 and RTX 4090, with input being a 1344×768 static image generated by Qwen-Image 2.1, and output being 226 frames (9.4s @ 24fps).

    Why it matters: Under the same seed, Pika H3 Flex's 6-step LCM schedule is about 4.8 times faster than 30-step base sampling; the image quality difference must be judged by yourself.

  5. Stuubzzz · 视频工作流实测75

    MiniMax H3 Character LoRA Training Test: 29 Clips, 4000 Steps, Why Step 3000 Was Chosen

    The author trained a MiniMax H3 character LoRA using AI Toolkit on an RTX 5090, with 29 3-second clips, rank 16, 4000 steps, taking about 15 hours of GPU time, and ultimately chose the step 3000 checkpoint.

    Why it matters: The author tested training a MiniMax H3 character LoRA using AI Toolkit on an RTX 5090, and provides the specific settings of 29 clips and 4000 steps, along with 5 reusable rules for reference when training other video models.

  6. Stuubzzz · 视频工作流实测70

    MiniMax H3 RefMod and reference image testing: long-video facial consistency comparison

    The author's testing found that MiniMax H3's RefMod can improve long-video facial matching from 0.23 to 0.80; the 0.5 MP version scored 0.75 in the second segment compared to 0.25 without RefMod.

    Why it matters: Testing shows that MiniMax H3's RefMod can improve long-video facial matching from 0.23 to 0.80; the 0.5 MP version offers better value.

Nov 1Sun
  1. Reddit StableDiffusion · 创作者与工作流分享26

    Krea 2 Fine-tuning Experiment: Surprisingly Effective, but Sacrificed Versatility

    The author performed seven consecutive fine-tuning runs on Krea 2, training on about 10,000 self-taken photos, resulting in a highly realistic photographic checkpoint, but at the cost of other styles and LoRA compatibility. The training set size was about 300 to 2,500 images, with up to 4,000 steps per run, and at checkpoint 12500 the photographic effect was excellent, but the model exhibited catastrophic forgetting. The author shares this experiment as a reminder that continuous full fine-tuning can lead to over-specialization of the model.

  2. Reddit comfyui · 创作与工作流分享53

    MiniMax H3 Singularity V2 New Workflow: Faster Motion, Better Faces, and Low VRAM Optimization

    Creator cgpixel23 shared an optimized ComfyUI workflow for MiniMax H3 Singularity V2, using INT8 quantization, Turbo LoRA, and optimized attention nodes to reduce VRAM requirements. The model is about 32GB. The new version has improvements over the previous version in fast motion, motion handling, distant faces, close-up portraits, visual style, prompt adherence, and audio reference processing. The author has publicly released the workflow and video tutorial.

  3. note · cocoredo_2509 · 图像视频创作分享56

    Reproducibility of AI production workflows: ComfyUI and multi-model collaboration

    The reproducibility of ComfyUI multi-model workflows hinges on synchronized management of production condition records and quality confirmation. Specific methods include evaluating characters and backgrounds separately, decomposing actions before generation, and comparing models under unified conditions. Costs should include labor and regeneration expenses, and it is recommended to start with small workflows and expand gradually.

  4. Reddit StableDiffusion · 创作者与工作流分享77

    Author creates 46-second sci-fi trailer entirely locally on MacBook M5 Max and reveals complete pipeline

    Author completes 46-second sci-fi trailer 'THE LAST SIGNAL' entirely locally on a MacBook Pro M5 Max 128GB, with no cloud rendering and no API calls, taking about 14.5 hours (including about 2 hours of waiting).

    Why it matters: The author reveals the complete local pipeline from storyboard to final film and the actual time taken, allowing creators to assess the feasibility of replicating a similar workflow on a MacBook.

  5. Reddit StableDiffusion · 创作者与工作流分享63

    Krea 2 LoKr Training Test: Testing clothing and character consistency across 10 concert eras at 1536px, and sharing LoKr vs LoRA comparison, training steps, and trigger word tips

    The author trained a 1536px Krea 2 LoKr to test clothing and character consistency across 10 concert eras, and after comparing with LoRA found that LoKr performed better. The optimal training steps were 1,500 steps (37.5 epochs), and using a real name as the trigger word worked better than a synthetic trigger word. The author also shared the ComfyUI workflow, LoKr weights, and 40 showcase images.

  6. Reddit comfyui · 创作与工作流分享45

    Complete music video with AI audio-visual and vocals overlaid, made with LTX 2.5 and ComfyUI

    The author used LTX 2.5 and ComfyUI to create a rap music video about GPUs, with all 52 clips generated locally. 13 Gemini still images each generated 4 LTX 2.5 image-to-video clips, batch-processed on a 5090 at 832x480 resolution, with each 4-second clip taking about 25 seconds. Beat-synced editing aligned the cuts to the drum hits, with hard cuts at vocal entries; the whole process took about 4 hours.

  7. Reddit comfyui · 创作与工作流分享70

    The author used MiniMax H3 and the self-developed Video Builder in ComfyUI to create a Halloween music video.

    The author used the self-developed Video Builder to have Claude automatically complete lyrics, prompts, and the render queue in ComfyUI, producing a MiniMax H3 Halloween music video. The entire video has 19 scenes, 2:29, with 80 renders in total and about 12 GPU hours; the author manually handled story ideas, the Suno song, and Nano Banana scenes, while Claude automatically generated characters, planned shots, and checked lip sync.

    Why it matters: The author published the complete division of labor and 2-pass workflow for Claude-driven ComfyUI music video production, which can be reused in one's own video generation pipeline.

Oct 31Sat
  1. FireCut · 口播剪辑与字幕教程17

    FireCut: 7 Quick Video Editing Tips Every Creator Should Know

    FireCut shares 7 tips for speeding up video editing, covering Premiere Pro's Q/W ripple trimming, Pancake editing, proxy editing, and adjustment layers, among others. In particular, FireCut uses AI in Premiere Pro and DaVinci Resolve to automatically remove silence and duplicate shots, generate subtitles and dynamic push-ins, and add B-roll and background music with one click (including the Storyblocks library).

  2. note · fordtaurus116 · 图像视频创作分享15

    SkyProduction: AI makes 3D spaces and camera movements freely available, using GPT Image 2.5 × Seedance 2.5 to actually produce videos

    SkyProduction combines models such as GPT Image 2.5 and Seedance 2.5 to provide a canvas, 3D space creation, and a camera movement library, allowing users to freely combine features to make videos. For 3D creation, you only need to paste in a background image and it automatically generates a 3D space; camera position and angle can be freely adjusted, or you can directly choose from the preset camera movement library. The experience is like being a film director, and the production process is free and fun.

  3. note · cocoredo_2509 · 图像视频创作分享68

    Pika H3 Flex: Import, Quality, and Cost Verification for Video Creators

    The author verifies Pika H3 Flex from a production-site perspective, pointing out that official performance cannot be directly equated with actual usability; character consistency, audio track separation, and 4K upscaling need to be evaluated separately. H3 Flex is based on MiniMax H3, supports up to 12 reference materials and 5 types of audio track separation, and after generating at 480p/768p can be upscaled to 1080p/2K/4K, but 4K is upscaled rather than natively generated.

    Why it matters: On-site verification of Pika H3 Flex shows that character consistency, audio track separation, and 4K upscaling need to be evaluated separately; the author's three short tests and recording method can be directly used for other model adoption decisions.

  4. Reddit comfyui · 创作与工作流分享58

    Creating a short animation scene from a character design sheet using FastH3 and ComfyUI

    The author used FastH3 and ComfyUI to create a ten-second animation scene from a character design sheet, and published the workflow, prompts, and reference assets. The workflow uses a custom Krea 2 Wulver fine-tune to generate the character design sheet, and FastH3 OmniRef generates a ten-second square output with eight-step sampling. Mochi was left out entirely. Custom model weights are not included, but the provided reference PNG can be used starting from the animation stage.

  5. Generative AI · 创作教程与项目分享44

    AI anime generators with fewer restrictions? A hands-on comparison of ChatGPT, Grok, and PixAI

    PixAI offers more control in anime image generation than ChatGPT and Grok; after testing the same prompt, the author found its style closer to anime scenes. PixAI has over 1.8 million community models and LoRAs, and its Tsubaki.3 model supports uploading three reference images to maintain character consistency, and can generate animations, stickers, or cards. ChatGPT Images 2.0 and Grok Imagine 2.0 excel at text-heavy images and speed, but style control depends on the prompt.

  6. Curious Refuge 创作教程69

    How to Use AI to Relight a Video (Step-by-Step Tutorial)

    Curious Refuge published a tutorial demonstrating how to use AI to relight videos, and tested the workflow's effectiveness on live-action footage and gray-box studio scenes. The workflow first exports still frames and relights them using an image model (Nano Banana 2.1 or GPT Image 2.5), then uses the relit images as references to apply to the original video. Testing found limitations such as exposure shifts, banding, and resolution loss, with output resolution lower than the 4K source.

    Why it matters: This workflow uses static-image relighting as a reference for the video model, allowing creators to evaluate the boundaries and limitations of AI relighting in post-production.

  7. 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.