Krea Video Unifies Veo 3.1, Sora 2 and Runway in One Browser
Krea bundles Sora 2, Veo 3.1, Seedance 2.5 and Kling 3.0 into one browser workspace with camera controls, presets and sound generation.
Sora
36 articles tagged with this keyword, sorted by most recent first.
I tried this Linux distro on a PC with just 3GB of RAM – and came away impressed
Essora Linux is a fast, minimalist Linux distribution designed for practical use on modest hardware.
Chinese video AI models beat US rivals in visual quality, but lag on safeguards: HKU study
Chinese AI video-generation models have surpassed leading US rivals in visual quality, but several fall behind in safety precautions, according to researchers at the University of Hong Kong. ByteDance’s Seedance 2.0 Pro took top spot in a ranking of 12 leading artificial intelligence video-generation models, scoring 3.84 out of 5 for visual quality. It outperformed Google DeepMind’s Veo 3.1, which scored 3.47, and OpenAI’s Sora 2 Pro, on 3.27, in a study led by Zhenhui Jack Jiang, professor of...
오픈AI, 비용 문제로 접었던 '소라' 외부 스타트업에 라이선스 추진
오픈AI가 높은 운영 비용과 컴퓨팅 자원 부담으로 개발 및 서비스를 중단한 비디오 생성 AI 모델 \'소라(Sora)\'를 외부 스타트업을 통해 다시 시장에 선보이는 방안을 추진하고 있다. 모델을 직접 가동하는 대신 외부 기업에 기술 접근권을 제공하고, 그 대가로 지분을 확보하는 방식이다.12일(현지시간) 디 인포메이션에 따르면, 할리우드 거물 제작자 제프리 카첸버그와 빌 피블스 전 오픈AI 소라 연구 책임자, 수제이 자스와 전 드롭박스 CFO가 공동 설립한 비공개 스타트업이 오픈AI와 소라 모델 활용을 위한 협상을 진행 중이다.이 스
South Korea struggles to curb sexual exploitation sites 10 years after Soranet scandal
A decade after the shutdown of Soranet – a sprawling online hub for secretly filmed and non-consensual sexual videos that became a symbol of South Korea’s digital sex-crime crisis – the country is still struggling to stamp out similar networks and copycat sites that have emerged in its wake. Their prevalence reflects the difficulties of deterring such illegal activities, policing the online space and changing societal attitudes towards women. Last month, a man was detained over an alleged...
Black Forest Labs' FLUX 3 Beats Sora 2 and Veo 3.1 on Physics
Black Forest Labs' new multimodal model tops Google DeepMind's physics benchmark for video generation, beating Sora 2, Veo 3.1 and Cosmos 3.
Comment on Which Agent Causes Task Failures and When?Researchers from PSU and Duke explores automated failure attribution of LLM Multi-Agent Systems by Nico
UPerNet's ability to simultaneously identify scenes, objects, materials, and textures through a hierarchical pyramid structure is brilliant. It reminds me of standard image processing workflows where you have to subtract before you parse. For instance, I'll typically use wordremover to quickly clear out unwanted text from an image, and then run deep visual analysis. It keeps the visual data much cleaner. Thanks for the detailed read!
Comment on NVIDIA’s Global Context ViT Achieves SOTA Performance on CV Tasks Without Expensive Computation by Sprite Creator
The article highlights how NVIDIA's GC ViT achieves SOTA performance on tasks like image classification and object detection without the usual quadratic compute cost, which is a huge step forward for practical deployment. I'm particularly interested in how the Global Token Generator efficiently captures long-range dependencies while keeping the architecture simple. This kind of efficiency could be a game-changer for real-time applications beyond standard benchmarks. https://spritegen.ai/
Comment on CMU & Inspired Cognition’s DocPrompting Improves Code Generation by Retrieving Relevant Documentation by SchemForge
The ability to update the documentation pool without costly retraining is a significant practical advantage. However, how does the system handle situations where multiple versions of the same API exist within the documentation, potentially leading to confusion? AI schematic creator
Comment on AI Video Generation Race Shifts from Capability to Profitability, Challenging Sora’s Dominance by digital mixtape
The capability to profitability shift is the real story here. Last year everyone was wowed by the demos, but wowed doesn't cover GPU bills, so this turn was inevitable tbh. Honestly think the efficiency point will decide it, since whoever can serve video cheap ends up owning the actual product layer. Side note, all this slick AI video oddly pushed me toward older formats. Needed to make a mixtape for a friend's birthday last weekend and didn't want to dig out my old drive, and a fixed tracklist felt way more personal than another endless autoplay feed.
Comment on 2020 in Review: 10 AI-Powered Tools Tackling COVID-19 by leony
Nice write-up. For restoring compressed or AI-generated clips, I've been trying https://flashvsr.com/ — cloud AI video super-resolution up to 4K.
Comment on Unveiling Sora: OpenAI’s Breakthrough in Text-to-Video Generation by Ella Fairweather
Yeah, it still been 2 years ago, but thanks check this virtual game , cheers.
World models that ignore human beliefs predict the wrong actions, new research shows
Current world models like Sora or Genie only simulate physics and ignore what people think, want, or feel. The new "Mental World Modeling" framework adds mental variables like beliefs and intentions. Even weaker language models using this approach outperform stronger models without mental modeling. The biggest bottleneck: predicting how physical and mental states change together. The article World models that ignore human beliefs predict the wrong actions, new research shows appeared first on The Decoder .
Italian town carries towering Virgin Mary structure through streets
Thousands gather in Sora, Italy, as devotees carry a 16-metre-high illuminated structure through the streets for the Feast of the Assumption.
Allow video creation (Sora) for Pro Users
The video production (Sora) was a very nice product, and I understand why it was discontinued. I think it would be very useful to add it back as an option for the tip paying pro tier. That would greatly reduce the millions of nonsense videos being made and allow the product to be used for business presentations and other professional purposes. It might also encourage more people to sign up for the top tier subscription. I know I used it for business presentations regularly and would love to see it back (even if it’s limited to the top tier).
Project SHAME - Sustainable Human Accountability Metrics Engine
AI Operating System for PC Shake the tree …That’s my strategy… AI will never die… If you’re still not following… oops. OK, so I learnt something today that genuinely surprised me: By 2023, annual global fossil-and-industry CO₂ emissions were roughly 65% higher than in 1993 — when I was 13, the age my son is now. Roughly two-thirds more CO₂ in a single year than when I was 13. Why? Consumption-based CO₂ per person Region 1993 2023 Change US 19.9 t 15.8 t −21% UK 11.4 t 7.1 t −38% EU-27 10.0 t 7.3 t −27% China 2.35 t 7.63 t +225% India 0.72 t 1.77 t +147% High-income countries 12.3 t 11.1 t −10% Upper-middle-income countries 2.38 t 5.31 t +123% World ~4.07 t ~4.62 t ~+13% Source: Our World in Data — Per capita consumption-based CO₂ emissions , using Global Carbon Budget data. Values rounded. Consumption-based accounting assigns CO₂ to where goods and services are ultimately consumed rather than simply where they are produced. The figures here concern fossil-fuel and industrial CO₂, excluding land-use change. So this does not mean the average person’s CO₂ footprint became 65% larger. Globally, consumption-based CO₂ per person increased much more modestly — by roughly 13% over this period. But the world’s population also increased enormously. Put the two together: The global arithmetic 1993 2023 Change World population ~5.59 billion ~8.1 billion ~+45% CO₂ per person ~4.07 t ~4.62 t ~+13% Global fossil/industry CO₂ ~22.8 Gt ~37.4 Gt ~+64% Approximately: 1.45 × 1.13 ≈ 1.64 This is…
GPT-Image - General AI Gallery: A Place for Images Created by Users
Secrets of a Desk
Feature Request: Native Video Generation & Editing in ChatGPT as a Premium/Pro Capability
I found and supported the previous community request regarding restoring and improving Sora/video access within ChatGPT. That discussion was closed on June 15, 2026, so I would like to expand on the concept with a broader proposal focused on native conversational video generation, video editing, organization, and premium-tier integration directly within ChatGPT. The Core Idea I believe ChatGPT should eventually support video in much the same way users can currently work conversationally with images. Instead of requiring users to leave ChatGPT, open another application, learn another interface, transfer their files, and continue their work elsewhere, a user could simply upload or generate a video and tell ChatGPT what they want done. For example: “Create a video from this concept.” “Make this scene longer.” “Remove this object from the video.” “Change the background.” “Turn these images into a video.” “Use this image as the character reference.” “Create a shorter version for social media.” “Add captions.” “Improve the video quality.” “Change the camera angle or movement.” “Continue this scene for another ten seconds.” “Create three variations of this scene.” The important feature would be the conversational interface . Users should not necessarily have to become professional video editors to perform useful video-editing tasks. Potential Capabilities A native ChatGPT video environment could eventually include: Text-to-video generation Image-to-video generation Video-to-video t…
AI video market has bounced back from Sora's false start
AI production companies like Promise are setting up shop around Hollywood's historic studios, using real-time backgrounds and other AI tools to cut film costs. Netflix already uses AI in 300 of its 1,000 titles, and the startup Higgsfield now carries a $5.4 billion valuation. What was once a tech demo has grown into its own industry, complete with valuations, job titles, and fights over who gets a cut. The article AI video market has bounced back from Sora's false start appeared first on The Decoder .
Sora 2 api stuck in in_progress state
Hi there — we’re closing this topic because there hasn’t been a new response for a long period of time and the discussion appears inactive. If you still need help, please start a new topic with your current question and any relevant, non-sensitive details. Thanks for being part of the community. Avinash
Blocked for Moderation Sora 2 API
Hi, My app allows users to use Sora 2 through the API. About 10% of requests get flagged for moderation. If OpenAI flags that request as moderation, does it still charge me usage? This is important because it determines if I should grant a refund to the customer or not. Thanks
Why are support cases and feedback threads being closed?
100 percent. No reason they can not at least provide updates on a timely basis. This company is way too rich to offer this poor support. I found a related thread from this month where the same thing was said. They basically called the person entitled for wanting a response within a week or two for what they thought was a critical issue. A company should be honest about it’s wait times and lack of support upfront. You have to be careful on these forums because any feedback perceived negative signals the moderators to close the thread , delete the post or ban.
Release Sora 2 open weights before shutdown
Welcome to the community! I’ve seen some really impressive work with open weight image and video generation recently. Of course, Sora and Sora 2 were incredibly impressive when they came out, and it’s always a shame when a model is lost (I’m still crying for gpt4-0314 and gpt 4.5), but I do suspect sora has been superseded at this point. Have you tried alternative models?
Comment on AI Video Generation Race Shifts from Capability to Profitability, Challenging Sora’s Dominance by The Choicer Voicer
It's interesting to see how the focus in AI video generation is shifting from pure capability to profitability. Makes sense that the market would mature that way. The Choicer Voicer
Feature Request: Built-in AI Video Creation Studio with Notifications in ChatGPT
Didn’t OpenAI stop developing SORA because it took up too much compute?
Comment on Microsoft’s Fully Pipelined Distributed Transformer Processes 16x Sequence Length with Extreme Hardware Efficiency by juego copero
Great article on FPDT! The memory hierarchy optimization is really impressive. As someone who runs a gaming site, I can appreciate efficiency improvements—smooth gameplay depends heavily on resource management. Check out https://juegocopero.com/ for more on that!
Sam Altman says he got ‘addicted’ to TikTok, including a Saturday afternoon where he scrolled ‘for like 3 hours’
OpenAI CEO Sam Altman said he got too sucked into TikTok when he forced himself to learn about the app when OpenAI was preparing to launch Sora.
CrossOracle: An Agentic Framework for Real-Time Expert Personas with Parallelized Acoustic Synthesis
Yash Raj Singh in Proceedings of the 27th Annual Meeting of the Special Interest Group on Discourse and Dialogue
Will there ever be a Sora 3?
Yes, There will ever be a Sora 3.
Sora 2 API jobs stuck at progress 0
Hey everyone, Is the API working as expected now, or is anyone still experiencing the issue? Avinash
Sora 2 all jobs stuck at 0% progress
For the whole day, all Sora-2 videos I’ve tried to generate through my app have been stuck at 0% with error=null. We use the OpenAi SDK. Why would this happen for all generations? I can’t even kill the jobs, they return an HTTP 400: “message”: “Video is still being processed”, “type”: “invalid_request_error” My longest job it’s been in_queue for almost 12 hours What am I doing wrong? Is the API failing? Is this related to the upcoming deprecation of Sora?
July 2026 (Theme: Perspective) — ChatGPT / API Image Generative Art Gallery, Prompt Tips, and Help
New theme: let’s make fair use in parody and enjoy the composition of many characters - an image completely within my power to have composed myself at significantly more work than “ask an AI”: Sgt Pepper cover - with a montage of cartoons. moderation_details: {"moderation_stage": "output", "categories": ["other"]} Then a prompt completely stripped of a single character’s name or a single intellectual property: denied again, against n>1. The only violation possible in this prompt is OpenAI violating intellectual property rights and then shifting the blame to the API “user error”. So, there is no fun image to show or share from OpenAI. You can paint your own picture: (big expand) Prompt and images from elsewhere (click for more details) IRL Daria vs Boris Badenough
Are video generation also great at next frame video prediction?
If I have a good video generation model like OpenAI's new Sora, will it be capable of doing just as well at next frame video prediction?