#Multimodal

AIEthan Mollick says he was wrong to think coordinating AI agents would require careful human-designed management structures. He points to personal agents like dots and Muse, and to a swarm of thousands of OpenAI agents that solved a Navier-Stokes problem in 88 hours with thin coordination. He argues many management problems stem from human limits, which agents lack, so people should mainly guide direction while agents handle organizing.

AIManus 2.0 introduces Video Editor, which lets users refine videos Manus generates, including changes to music, captions, and cut timing, without regenerating the entire video. The article describes Manus creating explainers, launch films, and animations from a single prompt, drawing on web search, video models such as Seedance 2.5, and code for motion graphics.

AISuno has launched Speech in beta, which it describes as the first audio model that generates voice and music together as one cohesive track. Users type an idea or existing text, then describe the voice and musical style they want, and the beta is now open to everyone after a month of testing with a small group.

AIGoogle DeepMind announced Gemini 4 Argon, rolling out first to trusted cyber defenders through its Fairwind Program. Argon will launch at an introductory price of $2 per million input tokens and $10 per million output tokens, with output limits raised to 1M tokens. The post cites a 77.9% score on DeepSWE v1.1 and 91.7% on LVBench, and says broad availability will follow safeguard testing.

Why it matters: The post pairs Argon's benchmark claims with the phased release, pricing, and safeguard details, helping readers weigh its frontier-level capabilities against its access limits.

AINVIDIA announced a livestream tomorrow at 9 a.m. PT in which it will build a visual AI agent from a single prompt. Viewers are invited to bring their questions to the session.

AILlamaIndex held Tuesday-night events in New York and San Francisco on document processing for AI agents, with the New York room filling a waitlist and San Francisco drawing almost 600 attendees. The talks focused on the problem that agents often receive document text without its layout, so they must guess which figures, such as a monthly rate versus a total on an invoice, mean what.

AIProject HydraFusion is now available in the GitHub Copilot app and VS Code, where users can select it like any other model. Behind the scenes, it routes a task across multiple models to draft, critique, revise, or escalate, then returns one combined result.

AIIdeogram 4.5 was built for precise, targeted editing but also performs very well across general editing tasks. Design Arena ranks it 15th in Image Editing with an Elo of 1250, placing it in the same performance band as MAI-Image-2.6 and Gemini 3 Pro Image Preview. It is especially strong at typography edits, such as modifying text in infographics.

AIIdeogram 4.5 is now available on Krea, and the post says it is built to handle repeated edits. The post states that repeated editing usually causes artifacts, pixel shifts, and color drift, which this model aims to solve.

AILucas Beyer says he could build a multimodal model that scores how well any freeform text fits a given image, with calibrated scores and sigmoid-based yes/no judgments. He floats raising roughly xxxM in funding, open-weighting the model, and keeping it under 1B parameters, perhaps around 400M.

AIIdeogram released Ideogram 4.5, which it calls the most precise edit model, claiming it avoids the artifacts, pixel shifts, and color changes that leading models add with each edit. The company says this eliminates artifact buildup and makes multi-turn editing possible. It is live in Ideogram, via the API, and with launch partners, with open weights promised soon.

AISenseTime previewed Dynamic Design, powered by SenseNova 6.8 Flash, which turns static images into animated visuals. The system decides which elements stay static and which to animate, chooses HTML/CSS, SVG, transparent images, or video for each element, and choreographs text reveals and subject motion. SenseNova 6.8 Flash is coming soon, and SenseTime is offering a limited beta.

AIKling AI showcased a short film generated by the full-powered KLING 4.0, which the source describes as the full-powered version. The background post from @hq4ai says the film was made with all-round reference generation, runs a native 30 seconds in 21:9 cinematic format, and offers clearer visuals and sound with more precise lip-sync than Flash. The same post states Kling 4.0 will launch in October.

AIInSpatio-World 1.5 from InSpatio_AI turns a single image, four images, a panorama, or a video into a navigable scene with wide viewpoint changes. The 1.3B model scores 68.72 on WorldScore-Dynamic, ranking first among evaluated real-time and interactive methods, with speeds up to 24 FPS. The post says the code is released under Apache 2.0 and that dependencies keep their own licenses.

AIKling 4.0 extends native single-pass video generation from 15 to 30 seconds and adds Multiple Keyframes supporting up to 10 keyframe images, versus Start & End Frames in Kling 3.0. It also expands reference inputs to up to 15 combined assets, including up to 5 videos totaling 30 seconds, and adds 10-bit HDR at 1080p and 4K. The all-new Kling 4.0 will officially launch in October, and Kling 4.0 Flash became available to a limited group of early-access users on September 28.

AIKling AI says Kling 4.0 will support native video generation from 3 to 30 seconds, with up to 10 keyframes for timing control. The full model is scheduled for October 2026, with early access to the new creation page rolling out from September 28.

AIJerry Liu benchmarked gpt-6.1 sol on document OCR tasks and found a sizable increase in table parsing and reading order over gpt-6 sol from a week earlier. Its table parsing is similar to gpt-6 astra. He noted frontier models still cost roughly an order of magnitude more than cost-effective document parsing solutions, leaving room to improve the premium end above 1c per page.

AIGoogle Research introduced Diffusion Controller, a framework that treats image generation as a continuous control problem rather than separate inference-time guidance and fine-tuning fixes. Its lightweight add-on "steering damper" network keeps the base model frozen and works on black-box or gray-box models, and it outperformed the industry standard on human preference matching. In a Stable Diffusion v1.4 test, the fully unlocked version achieved a 90% win rate over the baseline.

AIHiggsfield AI introduced Dots, a joint OpenAI x Higgsfield product offering an always-on creative crew that keeps working while users are away. Users can check in by text, call, or email and pause the work at any time. The post says it is powered by GPT-6.1 Sol.

AIMicrosoft Research has introduced Quine, an early-stage research effort to build a multimodal world model of biology that connects insights across biological scales and modalities. The system is designed to help scientists computationally search a space far larger than intuition allows and prioritize hypotheses before lab testing. Experimental results are meant to feed back into the model and sharpen future research directions.

AIDeedy describes a video generation pipeline built around Opus 5.5 in Claude Code, routing image, video, audio, and TTS models through OpenRouter's single API key. The workflow adds reference-image consistency, animatics before full renders, a critic skill that screenshots and transcribes output for QA, and ffmpeg for most editing.

AIShanghai AI Laboratory's Intern-Decision family of 0.8B, 2B, and 4B multimodal models averages 79.38, 84.68, and 90.02 across seven decision benchmarks. Intern-Decision-4B scores 88.74, surpassing Jev while achieving better probability calibration. Reported mean latency is 33.98, 33.28, and 44.16 ms, versus 109.70 ms for Jev in the same local HF setup.

AIOpenBMB announced that MiniCPM-o 4.5 is now supported in SGLang Omni v0.1.7, giving developers more flexibility to run and build with the model. The background release notes add that MiniCPM-o 4.5 brings multimodal input and speech output to the runtime. MiniCPM-o and MiniMax-Music3 also gained Intel XPU support in the same release.

AIA user ran qwen3-vl:32b-instruct locally through Ollama on a 128GB computer to auto-tag 40,000 images in their Eagle app. The image library was collected over 10 years and is intended as a personal asset base for Claude Code-made knowledge videos. The user says the use case makes the 128GB memory purchase feel worthwhile.

AIThe article presents a four-part prompt structure (action verb, target element, desired result, protection instructions) for AI photo editing, saying it preserves approved work across platforms. It identifies three common failure causes: unmatched light direction, stacked edits in one prompt, and vague visual language. It states that simple skin retouching takes 2-3 minutes versus 15-30 minutes manually.

AIWorld Labs is joining AMD, according to a post from AMD and World Labs' own announcement. World Labs says it aims to scale its efforts and reach while getting closer to hardware in pursuit of spatial and physical intelligence.

AILlamaIndex argues that frontier vision-language models still fail on real forms such as W-2s, 1040s, W-9s, and scanned W-4s, because forms require detecting every field, preserving section hierarchy, linking values to their exact boxes, and reading handwriting and checkmarks. The company's blog post details these failure modes and presents a custom cookbook for LlamaParse as a cheaper way to handle such forms.

AIGoogle says Gemini 3.8 Flash, its most intelligent workhorse model, improves on 3.7 Flash in software engineering, agentic tasks, and multistep reasoning by running extra reasoning steps and calling tools iteratively. The post highlights four community builds, including a model rocket simulation, an animated ink-painting effect, a 3D dinosaur skeleton, and an interactive automatic transmission simulation. Developers can try the model through Google Antigravity and Google AI Studio.

AIKling AI says its Kling 4.0 Flash is live now for Ultra Yearly subscribers, with the full Kling 4.0 coming this October. The update advertises up to 4K resolution, 10-bit HDR output, stereo audio, and native 30-second generation. It also adds Omni Reference supporting up to 15 multimodal references and multi-keyframe control with up to 10 keyframes.

AIBona Film Group announced that Sanxingdui: Future Past, a 100-minute film using AI throughout production, will screen nationwide on October 23, 2026. The production team says AI handled tasks like image generation while over 100 professionals kept creative control, and it took two years and over 1.2 million source images to maintain consistency across the film.

AIJina-OCR-v1, a 3.4B-parameter MoE model that activates 570M parameters per token, converts entire document pages into structured Markdown at 2.57 pages per second. It scores 91.14 on OmniDocBench v1.6 and 83.4 on olmOCR-Bench, 7.4 points above DeepSeek-OCR on the latter, and delivers the highest throughput among 14 evaluated systems at concurrency 32. The model is released under CC BY-NC 4.0, so commercial use requires permission.

AIDeedy argues every open source repository should have a roughly seven-minute explainer video like the one made for SQLite, covering its purpose, a high-level code map, a query's path through the codebase, core abstractions, and a real execution trace including join-order query planning. He says the video was generated with Opus 5.5 and Gemini 3.8 TTS, and he expresses amazement at how coherent and capable the model is.

AIAnthropic's Felix Rieseberg had Opus 5.5 create a parody of the "nihilistic" penguin, generating all models and textures from scratch. The model did this by using Blender as a Python library.

AIDeepMind has published an essay arguing that AGI will emerge through "cooperative interactions among models, tools, institutions, and human participants" rather than from a single winning AI. The commentary supports the collective framing but rejects treating AI agents as having their own theory of mind, arguing that creating new moral subjects should remain humanity's remit.

AIShanghai AI Lab's InternLM released Intern-Decision-2B, a multimodal structured decision model fine-tuned from Qwen3.5-2B that returns calibrated answer distributions for multiple questions in one forward pass. It averages 84.68 across listed benchmarks with a 0.437 Brier score and 33.28 ms mean latency on a single RTX 4090. Model weights, a Python DecisionEngine API, and GitHub code are available, with support for up to 16 questions and eight images.

AIGoogle AI's weekly roundup lists Gemini 3.8 Flash TTS and Flash-Lite TTS as expressive audio generation models. It also announces Gemini 3.8 Live with Live Avatar for near real-time visual conversation and a Live Chat voice feature on the Gemini Notebook mobile app across about 100 languages. Project Suncatcher will launch a prototype satellite to test Google TPUs in orbit and explore solar-powered AI compute in space.

AIClaude Tag in Slack can now use users' personal connectors, such as Drive, Salesforce, and warehouse access, within channels. The author says Tag writes over 50% of their PRs daily and handles nearly all their data analysis and many product bug fixes. The personal connector feature is available on Teams today and Enterprise next week.

AIAnthropic's Noah Zweben shared a claymation video showing Claude Code's /remote-control feature, now made with Opus 5.5 after an earlier Opus 4.6 version. The feature, which lets users control Claude Code remotely, is rolling out to Pro users at 10% and ramping, with Team and Enterprise support coming later.

AIMeituan's LongCat team has released LongCat-2.5-Preview, a natively multimodal model with 1.6T total parameters, about 48B active, and a 1M-token context window. The model is built for long-horizon tasks spanning terminals, browsers, GUIs, spreadsheets, and design tools. It is available now through an API on the LongCat platform and a chat interface.