Skip to content
Date and source

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

Oct 18Sun
  1. Stuubzzz · 视频工作流实测66

    Hands-on test of Qwen-Image 2.1 Turbo in ComfyUI: 8-step scheduling and 300 renders

    The author tested Qwen-Image 2.1 Turbo with int8 files on an RTX 5090: native 2K (4 MP) images averaged 11.8 seconds, versus 36.9 seconds for the 30-step base model, about 3.1 times faster.

    Why it matters: The author used 300 renders to test the differences between Turbo and the base model, and provides reusable 8-step sigma and resolution settings.

  2. note · fordtaurus116 · 图像视频创作分享36

    Vidu Q4 Preview hands-on: making Japanese character animation with GPT Image 2.5 × Pollo AI

    The creator tested the video generation model Vidu Q4 Preview on Pollo AI, generating original character illustrations with GPT Image 2.5, then produced a 5-second Halloween short and a 16-second self-introduction animation, with natural Japanese dialogue intonation and lip sync. The model has a low cost for short animation production, and the credits required vary with video length and resolution.

  3. Reddit StableDiffusion · 创作者与工作流分享20

    Direct same-prompt comparison of MOSS SFT, Flux 3 Klein Distilled, and the original SD 3.0

    A creator directly compared MOSS SFT, Flux 3 Klein Distilled, and the original SD 3.0 with the same set of prompts, concluding that SD3 is completely on a different level from any recent model. There were 4 test prompts in total, covering scenarios such as high-fidelity social media portraits, 3D game screenshots, cafe architectural photography, and surveillance camera footage.

  4. Reddit StableDiffusion · 创作者与工作流分享22

    Same-prompt comparison of anime scene generation with SDXL (WAI-Illustrious v17), Krea 2 turbo, and MOSS-Image-SFT

    A user compared anime scene generation across three models—SDXL (WAI-Illustrious v17), Krea 2 turbo, and MOSS-Image-SFT—at 1024×1024 using the same set of pure booru tag prompts (no style prefix, no quality words, no negative prompts), with 4 scenes × 2 seeds, all without LoRA.

  5. Monster社区 · 中文ComfyUI实测与工作流38

    FLUX.3 Klein Three-Model Hands-On Test: Prompt Constraints and Continuous Editing

    The author used 35 groups and 103 test images in ComfyUI to compare FLUX.3 Klein Q8, FLUX.2 Klein BF16, and Qwen-Image 2.1 INT8, checking portraits, Chinese packaging, quantity and spatial relationships, and continuous editing. In Chinese and Chinese-English packaging samples, Qwen-Image 2.1 better met the requirements; obvious problems appeared in the later stages of continuous editing, and only FLUX.2 completed the final background modification. The author believes FLUX.3 does not outperform the previous generation in all scenarios and still has shortcomings in Chinese.

  6. Wireflow · AI短剧与视频制作教程36

    Higgsfield Ad Multiplier Alternatives: 8 Ways to Replace Ad Products or Actors (2026)

    Wireflow used the Kling O3 Edit Video node to complete an ad product replacement test: replacing a black beverage can with a terracotta mug in a 7-second 720p vertical ad, taking 3.5 minutes and consuming 142 credits (approximately $1.49–$2.13 depending on the plan). The person, action, background, and audio all remained consistent, but the mug retained the can's taller silhouette.

Oct 17Sat
  1. Generative AI · 创作教程与项目分享34

    Tensor.Art vs. PixAI anime AI generators: which should you choose?

    A hands-on comparison of two browser-based anime AI generators, Tensor.Art and PixAI, testing text scenes, signboard text, and a crowded four-person scene with three prompts. Tensor.Art requires choosing checkpoints, stacking LoRAs, and tuning parameters yourself, and supports ComfyUI workflows and self-trained models; PixAI focuses on its in-house model Tsubaki.3, allows full-sentence descriptions, character reuse via @image1, and instruction-based image edits, and hosts over 1.8 million community models and LoRAs.

  2. Frank Houbre · AI电影制作实战50

    Higgsfield 2026 hands-on: what one director kept and what he threw away

    Director Frank Houbre tested Higgsfield 2026: one account can call more than twenty models including Seedance 2.5, Kling 3.0, Veo 3.1, and Sora 2; Cinema Studio 4.0 is based on Seedance 2.5 and provides Director's Panel parameters such as camera, lens, and aperture, with up to 50 references and 30-second clips per generation.

  3. 我爱计算机视觉 · 公开公众号 RSS38

    Tencent Photon x NUS Release Orbia Video World Model Evaluation Benchmark

    Tencent Photon and NUS propose the Orbia video world model evaluation benchmark, using 720 exploration and revisit cases to test 27 models, checking retention of known scenes, persistence of newly generated content, and cross-view 3D consistency. SolarWM-H3 ranks first overall with 76.32, followed by Lyra2 and EVOKE in second and third place; content persistence in long video generation is generally lower than in short videos, and the average score for forward-backward trajectories in the world model group drops from 66.21 to 40.96.

  4. Reddit StableDiffusion · 创作者与工作流分享66

    Hands-On with MOSS-Image 5B INT8 Quantization and 8-Step DMD on a 12 GB RTX 3060

    The author tested MOSS-Image (5B single-stream DiT, Qwen3.5-9B text encoder, FLUX.2 VAE, Apache 2.0) on a 12 GB RTX 3060 using 35 difficult prompts and a single seed (42), without cherry-picking images and including failure cases.

    Why it matters: A hands-on test of MOSS-Image's INT8 quantization and 8-step DMD acceleration on 12 GB of VRAM, with a full comparison of text rendering, layout and failure cases.

Oct 16Fri
  1. MindStudio Blog · 创作分享56

    Krea 3 Hands-On Test: An AI Image Editor Focused on Speed, Not Generation

    Krea AI released Krea 3, only five months after Krea 2, shifting its positioning from generation to image editing; the company calls it the fastest option for 2K editing. The core new feature, precise edit mode, lets you box-select a region on an image and change only that part; in testing it completed outfit-change edits that Nano Banana and GPT image tools refused on grounds of inappropriate content, and it also performed well at removing grain and noise from GPT-generated images and improving skin texture.

  2. 新智元 · 公开公众号 RSS71

    Shengshu Technology Releases Vidu Q4 Preview, with Launch Pricing as Low as 0.09 Yuan/Second

    Shengshu Technology released Vidu Q4 Preview, focused on character performance, camera movement, and complex visual effects, strengthening character acting and dynamic camera work in 2D and 3D anime scenes, with launch pricing starting as low as 0.09 yuan per second.

    Why it matters: Vidu Q4 Preview bundles character performance, camera movement, and multi-reference capabilities and offers launch pricing starting at 0.09 yuan per second, which can be used to judge whether test-shoot costs have entered an affordable range.

  3. Reddit StableDiffusion · 创作者与工作流分享22

    Comparison of FLUX 3 Klein, FLUX.2 Klein 9B, Qwen Image 2.1 and Krea2

    A user compared four models—FLUX 3 Klein, FLUX.2 Klein 9B, Qwen Image 2.1 and Krea2—and found that Krea2 and Qwen Image 2.1 clearly lead the FLUX series in human anatomy and realism, with FLUX 3 performing worst on human anatomy. FLUX 3's prompt-following ability is still considered the best, but the author believes its capabilities may be weakened by excessive and unnecessary censorship.

  4. Reddit StableDiffusion · 创作者与工作流分享22

    Comparison of basic editing between Flux.3 klein (distilled version) and Qwen image 2.1 (turbo)

    A user compared the image editing speed of Flux.3 klein (distilled version) and Qwen image 2.1 (turbo) on a 4070s: the former takes 10 seconds for 1mp with euler 16 steps, while the latter takes only 3 seconds with euler 8 steps. The test covered prompts such as adding a robot dog, recomposing, changing lighting, converting to black-and-white comic, expression sheet, clothing variants, and figurine transformation.

  5. Reddit StableDiffusion · 创作者与工作流分享22

    Minimax H3 vs. Seedance 2.0 camera movement comparison test

    A user compared the multi-camera movement performance of Minimax H3 and Seedance 2.0 using the same 15-second martial arts training prompt. The prompt required presentation in the style of a 1986 American martial arts film, including continuous camera movements such as dolly-in, lateral tracking, 180-degree orbit, and low-angle tracking, while keeping the character's face, clothing, and body type consistent. The test did not give a clear conclusion on which was better.

  6. Fstoppers60

    Is AI color grading good enough? Fstoppers tests Imagen Video

    Fstoppers author Patrick Hall tested Imagen Video's automatic color grading capability with two real projects. In a multi-camera commercial test, Imagen completed in about 35 minutes color grading that would originally take several days of manual work, and about 20% of the clips still required manual adjustment; in a three-camera podcast project, after the update it only processed materials on the timeline, reducing the time from nearly 4 hours to under 40 minutes.

  7. Reddit StableDiffusion · 创作者与工作流分享22

    Flux 3 Klein performs acceptably on lighting and atmosphere, but details break down on biological subjects

    After hands-on testing of Flux 3 Klein, a user found that the model performs better than most models on atmospheric scenes such as snowstorms, but its details are noticeably chaotic compared with other current models, and proportions are severely distorted when biological subjects are involved.

  8. 娱乐资本论 · 澎湃商业与影视趋势70

    娱乐资本论's comparative review of 10 AI video canvas platforms: functions diverge, with different emphases for three user groups.

    娱乐资本论 releases the functional review of AI video canvas, testing 10 platforms (AniShort, 巨日禄, LibTV, 纳逗Pro, OiiOii, Seko, ShotLab, TapNow, 万兴剧厂, 小云雀), and organizes functional differences according to three user categories: serialized comic dramas, professional film/TV, and cross-border e-commerce.

    Why it matters: The review covers 10 AI video canvas platforms, sorting out functional differences for three user categories—comic dramas, film/TV, and e-commerce—to help creators choose according to their needs.

  9. Higgsfield 官方博客38

    7 best Veo alternatives: how to maintain consistency during generation

    The official Higgsfield blog compared 7 Veo alternative tools, focusing on the consistency of characters, clothing, and scene details across multiple shots. Cinema Studio 4.0 supports up to 50 reference assets, 30-second clips, and 1080p output, with a 10-second 720p clip including audio consuming 70 credits; Seedance 2.5 supports 30-second native stereo and 50 references, and Kling 3.0 can generate up to 6 shots at a time.

  10. Reddit StableDiffusion · 创作者与工作流分享18

    Krea 3 vs Qwen 2.1 vs ZIT: a random comparison of text-to-image models

    A user on Reddit conducted a random comparison test of three text-to-image models: Krea 3, Qwen 2.1, and ZIT. The user believes Krea 3 is better than the other models at understanding subtle facial expressions such as biting one's lower lip, but it can also generate unsettling images from simple prompts, which stand out even more because of their realism.

  11. note · miyuu_2026 · 图像视频创作分享22

    Lakeside girl generation test: English vs. Japanese prompt comparison across Z-Image-Turbo, Flux.2-Klein, Qwen Image 2.1, Krea-2 Turbo, and Flux1-dev

    The author used the same set of lakeside girl prompts in ComfyUI to compare English and Japanese generation results across five models: Z-Image-Turbo, Flux.2-Klein 9B, Qwen Image 2.1, Krea-2 Turbo, and Flux1-dev. Overall, English prompts produced more realistic results, Krea-2 Turbo performed best, Qwen Image 2.1 exhibited obvious AI faces with poor results, and Flux1-dev could not generate with Japanese prompts.

Oct 15Thu
  1. 沃垠AI · 公众号公开订阅62

    CapCut launches new AI features including CapCut Assistant and CapCut Hub

    At its AI new creation launch event in late September, CapCut comprehensively upgraded the PC version and APP, launching CapCut Assistant (called Xiaoying on the APP) and the canvas-based product CapCut Hub. The author tested using CapCut Assistant to edit 8 video clips into a vlog with voiceover, subtitles, transitions, and filters, taking about 5 minutes in total, and said that repetitive operations such as trimming voiceover scripts, slicing long videos, and commentary videos can all be handed over to it.

  2. 娱乐资本论 · 搜狐公开专栏62

    AI video canvas feature comparison: 10 platforms diverge by comic drama, film and TV, and e-commerce going global

    Entertainment Capital Review conducted a feature comparison of 10 AI video canvases—AniShort, Jurilu, LibTV, Nadu Pro, OiiOii, Seko, ShotLab, TapNow, Wondershare Drama Factory, and Xiaoyunque—with data as of October 5, 2026. The review organizes each platform's focus according to three user types: comic drama serials, professional film and TV, and e-commerce going global, and lists the standard features all 10 have as well as the exclusive capabilities held by only one or two. Similar features such as consistency anti-face-swapping, script storyboarding, audio-video separation, and team collaboration are also compared item by item.

  3. note · pixolm_ai · 图像视频创作分享22

    6 Halloween AI Image Prompts: Photo Processing and Icon Creation

    Pixolm used Nano Banana 2.1 under Custom Create with 1K and Standard settings to generate 1 image each, completing 6 Halloween examples: 3 photo-processing images based on the same fictional adult female reference image, and 3 icons generated purely from text, at 8 Credits per image and 56 Credits in total. The photo processing preserved the person's facial features and 4:5 composition, replacing only clothing and background; the icons used no character reference, instead specifying the character, color scheme, and circular negative space.

  4. GitHub:AI视频开源项目发现38

    StreamEval Releases an Interactive Streaming Video Generation Evaluation Benchmark

    StreamEval is the first evaluation benchmark for interactive streaming video generation, covering sessions from 15 to 300 seconds and testing a system's ability to respond to new requests and maintain context during video generation. The benchmark includes five tasks—digital human dialogue, virtual human performance, world generation, instruction editing, and reference editing—and ten metrics, each scored on a 0–100 scale, with no combined total score across tasks. The data, evaluation code, protocol, and scoring rules are all public, and the models and leaderboard are updated monthly.

  5. note · apt_zebra8914 · 图像视频创作分享22

    ChatGPT vs. Gemini image generation comparison: 5 rounds of testing with the same prompts, Gemini scores 70 vs. ChatGPT's 65

    Under the same set of prompts, ChatGPT (Pro, thinking level "high") and Gemini (Nano Banana 2.1) scored 65 and 70 respectively across five tests: thumbnails, photorealistic style, funny comics, posters, and partial edits. ChatGPT tends to "polish results to look better," deviating from the specified style in photorealistic and comic categories; Gemini is more faithful to the prompts, but the emphasis treatment of the word "そう" in the poster was penalized. Both scored full marks on partial edits.