Skip to content

Content Formats

Model Releases Latest

Video/Image/Audio Generation Model Releases, Open-Sourcing, and Iterations: First-hand coverage of flagship updates, open APIs, and performance and pricing changes.

123 picksLast 30 days 45 itemsCollected 367 items

updated

Picks for Model Releases

TodayNov 19Thu1–20
  1. Google DeepMind68

    Genie 4 is now available.

    Genie 4 is now available to Google AI Ultra subscribers, supporting real-time world editing during sessions and real-time native audio. Genie 3 has been updated in sync, with single generations up to 3 minutes long, supporting 360-degree panorama seeds and real-time world events; new Project Genie agent and community gallery have been added.

    Why it matters: Genie 4 is available to Google AI Ultra subscribers, supporting real-time world editing during sessions and real-time native audio, allowing creators to evaluate its interactive world generation, sharing, and re-creation capabilities.

Nov 18Wed
  1. 爱范儿65

    Jimeng Seedance 2.5 launches a 4K version, supporting up to 30 seconds of single generation and six aspect ratios.

    Jimeng Seedance 2.5 launches a 4K version, supporting up to 30 seconds of single generation, full-modal reference, video editing and extension, and adopts 10-bit BT.709 color format and six aspect ratio outputs.

    Why it matters: The author tested Seedance 2.5 4K in four scenarios: brand TVC, e-commerce, film/TV previsualization, and large-screen display, demonstrating the practical value of 4K quality for commercial production workflows.

  2. 可灵官方博客65

    Kling releases Kling Image 4.0, focusing on realism and native 8K generation.

    Kling has released its next-generation image generation model, Kling Image 4.0, now available on Kling Creative Studio and open to enterprises and developers via API. The model is based on a native unified understanding and generation architecture, supporting native 8K image generation, text rendering of up to 2000 characters per image, generation of up to 10 consistent-style images from a single prompt, and long-image generation such as e-commerce detail pages. It also offers 110+ preset color grading styles, 70+ preset commercial fonts, and support for seven languages.

    Why it matters: The new model unifies visual understanding and image generation in the same autoregressive framework, and its native 8K and long-text rendering capabilities are worth creators evaluating as a replacement for existing image generation workflows.

Nov 16Mon
  1. Zyphra · Zonos 语音模型66

    Zyphra releases ZONOS3-Mini multilingual TTS model

    Zyphra releases compact multilingual TTS model ZONOS3-Mini, supporting 43 languages, with approximately 100 million active parameters per token, and provides a native transformers port. The model uses ZAC-1 codec as audio tokenizer, includes 27 built-in release voices, supports voice cloning from short audio, and is licensed under Apache-2.0.

    Why it matters: ZONOS3-Mini is released as a native transformers port, supporting 43 languages and voice cloning, facilitating evaluation of its suitability as a multilingual TTS baseline.

Nov 14Sat
  1. 钛媒体67

    MiniMax H3-Image Hands-On Comparison Against GPT Image 2.6 and Seedream 5.5: Faster Speed, but Unstable Execution of Complex Instructions

    MiniMax released the H3-Image model weights on November 10. The company says it can generate a 16-megapixel high-resolution image in 10 seconds, at a price about 50%-70% of mainstream tools in the same tier.

    Why it matters: The three new models show clear ranking gaps on text-to-image and image editing leaderboards. Hands-on testing shows MiniMax generates images fastest but is unstable in executing complex instructions, while OpenAI leads overall.

Nov 13Fri
  1. 美团 LongCat · 创作模型70

    Meituan LongCat releases LongCat-Video-Avatar 1.6, an audio-driven human video generation model.

    Meituan LongCat releases the open-source audio-driven human video generation framework LongCat-Video-Avatar 1.6, which improves lip-sync, coordinated facial and body movements, and long-video consistency compared to 1.5, supports video generation over 10 minutes, and optimizes inference speed.

    Why it matters: The new version improves long-video stability and inference efficiency, and in human preference evaluation it ranks second only to OmniHuman-2.0, providing a reference for selecting digital human video generation solutions.

Nov 12Thu
  1. RenderBob · 图像视频工作流72

    Seedance 2.5 Flash released, priced at approximately 0.99 yuan/second, offering a cheaper and faster tier

    Seedance 2.5 Flash was released on November 12. A 5-second 720p no-input video example is priced at approximately 0.99 yuan/second, about 34% cheaper than Seedance 2.5 at the same specifications, but 1080p and 4K are not yet supported. The vendor claims generation speed is about twice that of Seedance 2.5; the author notes this data comes from the vendor and no independent benchmark has been seen. The API is expected to launch on Volcano Engine on November 15, and will gradually become available in applications such as Jimeng.

    Why it matters: Seedance 2.5 Flash is priced about 34% lower than Seedance 2.5, and speed is claimed to be about 2 times faster, but the author reminds that these data come from the vendor and need to be verified independently.

  2. 量子位 · 搜狐公开专栏71

    MiniMax open-sources the H3-Image image model, supporting native 4K and up to 16 reference images

    MiniMax has open-sourced the H3-Image image generation model, which supports 16MP (4096×4096) native 4K output and up to 16 reference images as input. In the author's hands-on testing, it can generate complex images such as CAD three-view drawings, mechanical exploded views, and advanced calculus exam papers, with web-based 4K generation taking about 10–20 seconds. API pricing is 50% to 70% of mainstream competitors, and 16MP requests receive an additional 50% discount during the first week.

    Why it matters: MiniMax has open-sourced the H3-Image image model, supporting native 4K and up to 16 reference images. The author's hands-on testing covers its CAD breakdowns, complex layouts, and the usability of its low-priced API.

  3. 站长之家 · AI影视与短剧产业75

    ByteDance releases Seedance 2.5 flash: generation speed doubles, 720P generation at about 0.99 yuan/second.

    ByteDance Volcano Engine releases Seedance 2.5 flash text-to-video model, with generation speed doubled compared to Seedance 2.5, and 720P generation price at about 0.99 yuan/second, down about 34% from Seedance 2.5. The model supports up to 30 seconds per segment, offers two resolutions: 480P and 720P, and the API will be available on Volcano Ark on November 15.

    Why it matters: Seedance 2.5 flash enters batch production scenarios at a 720P generation price of about 0.99 yuan/second, offering a cost-effective solution for high-frequency content production.

Nov 11Wed
  1. Reddit StableDiffusion · 创作者与工作流分享75

    PAN 2 released: open-source 29B video generation model

    PAN 2 released, an open-source 29B latent diffusion model supporting video generation up to 20 seconds, 720p, 24fps, under Apache 2.0 license. The publisher also released training code and provides integrations with Diffusers, FastVideo, and vLLM. The model enables an LLM to generate dense text states/scripts, timing events and actions, with camera nested under events but decoupled from actions.

    Why it matters: Open-source 29B video generation model and training code released, allowing creators to evaluate its 20-second 720p generation capability and LLM-driven script-level controllability.

Nov 10Tue
  1. Zyphra · Zonos 语音模型65

    Zyphra releases open-source speech model ZONOS3-Mini

    Zyphra releases open-source compact speech model ZONOS3-Mini, with approximately 100M active parameters, capable of real-time operation on phones, laptops, and browsers. The model supports cloning voices from audio samples, designing characters by age/accent/timbre, and word-by-word control of emotion, speaking rate, and expression. Weights are open-sourced under Apache 2.0, and it can be served via SGLang-Omni on NVIDIA GPUs with an OpenAI-compatible interface.

    Why it matters: With 100M active parameters, it can synthesize speech in real time in the browser, providing a locally runnable open-source solution for localized dubbing and character voices, supporting voice cloning and word-by-word emotion control.

Nov 9Mon
  1. 字节 Seed 官方博客63

    Seedream 5.5 released: supports 100 rounds of stable editing and native RGBA output.

    ByteDance released the next-generation image generation model Seedream 5.5, which supports 100 rounds of multi-round editing, native RGBA transparent background output, and improved image realism and text rendering. Compared with Seedream 5.0 Pro, the usability rate across six production scenarios has improved, with film/TV and office scenarios exceeding twice the previous version; the severe text corruption rate dropped to 0.4%.

    Why it matters: Seedream 5.5 focuses on 100 rounds of stable editing and native RGBA output. Evaluations show the severe text corruption rate dropped to 0.4%, providing more reliable image assets for professional visual production.

Nov 7Sat
  1. z-ab · 开源短剧与创作项目分享74

    OpenBMB open-sources VoxCPM2: 30 languages, 48kHz audio quality, voice design and cloning completely free

    Tsinghua OpenBMB team open-sources VoxCPM2 speech synthesis model, 2 billion parameters, supports 30 languages and 48kHz audio quality, free for commercial use under Apache-2.0 license. The model uses a diffusion autoregressive zero-tokenizer architecture, supports natural language voice design, controllable cloning and ultimate cloning, and achieves English WER 1.84% and Chinese CER 0.97% on the Seed-TTS-eval benchmark.

    Why it matters: Open-source TTS model VoxCPM2 is free for commercial use under Apache-2.0 license, supports 30 languages and 48kHz audio quality, creators can evaluate whether to replace existing paid voice-over solutions.

Nov 6Fri
  1. 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. Oimi · 中文图像视频教程75

    GPT-Image-2.6 Release: New Features, Flare vs. Sunburst Comparison, and Full API Pricing Breakdown

    OpenAI released GPT-Image-2.6 on November 4, 2026, with generation speed improved by about 50%, and Flare can generate images in as little as 6 seconds. New features include GIF generation, reusable reference characters, transparent PNG annotation editing, and SVG export. API pricing is the same as 2.5, but output token usage for low/medium tiers may increase, so real-world testing is needed before migration.

    Why it matters: GPT-Image-2.6 adds GIF generation and reference characters. API pricing is the same as 2.5, but output token usage may increase, so real-world testing is needed before migration.

  2. ComfyUI Wiki · 工作流与开源创作分享77

    Viggle Remake Release: Open-Source Video Remake Model Based on MiniMax H3

    Viggle releases open-source video remake model Viggle Remake, based on MiniMax H3 base transformer, offering two rank-256 LoRAs for swap and edit, as well as ComfyUI node package, int8/GGUF quantized builds, and MCP server.

    Why it matters: Viggle Remake open-sources video remake capability to ComfyUI, offering swap and edit modes and quantized versions, allowing creators to replace characters or objects in footage.

Nov 3Tue
  1. 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.

  2. ComfyUI Wiki · 工作流与开源创作分享68

    TaoLive AIGC releases TaoMate-FVSR, a video face restoration and visual dubbing model.

    On November 2, TaoLive AIGC released TaoMate-FVSR, a 10.3 GB video face restoration model that can fix blurry faces and fused teeth in generated videos, and make lip movements match new speech when replacing audio tracks.

    Why it matters: Addressing the defects of blurry faces, fused teeth, and lip movements not matching the audio in generated videos, it simultaneously performs restoration and lip-syncing, and can be run directly in ComfyUI.

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. ComfyUI Wiki · 工作流与开源创作分享83

    Pika H3 Flex Release: Six-Step Video Generation Model with Open-Source Weights and ComfyUI Template

    Pika releases H3 Flex, which post-trains and distills MiniMax H3 to six-step sampling, with open-source weights and ComfyUI template. The model loads onto the H3 base with two LoRA adapters, supporting text-to-video, first/last frame, and reference-condition generation, outputting 480p/768p, 5-15 seconds, and mixing in 32kHz stereo audio. Pika's self-evaluation shows it leads comparison models on five visual and audio quality metrics; community testing suggests high-frequency detail flickers at four steps, recommending keeping six steps.

    Why it matters: Pika post-trains MiniMax H3 into a six-step model and open-sources the weights. The ComfyUI template can be loaded directly, allowing creators to evaluate the quality of six-step output accordingly.