Updated
Updated
Showing low-relevance items too. Hide low-relevance items
OpenRouter@OpenRouterOfficialAI score34AIElevenLabs is now available on OpenRouter, offering 9 Text to Speech models supporting up to 90+ languages and 2 Speech to Text models with speaker labels and word-level timestamps. All ElevenLabs models are 50% off on OpenRouter through October 19, 8am PT.
LangChain@LangChainOfficialAI score22AILangChain announces LangSmith Engine v2, which red teams agents, detects more issue types, and automatically tests proposed fixes. The update was presented by Ben Tannyhill at the Interrupt NYC session.
elvis@omarsar0XAI score44AINVIDIA's VERA turns benchmark trajectories into over 9,000 restartable sandboxes with rubric scoring and updates both model weights and the agent harness together. A harness edit is kept only if it adds at least 5 points on the development set, and a checkpoint is rejected if its score drops more than 20%. At 27B, the co-evolved agent scores 71.6 on AutoCoWorkBench, above Claude Opus 4.8, and the environment corpus is open-sourced.
Google Gemma@googlegemmaOfficialAI score41AIGoogle's Gemma account promotes research showing how AI can help scientists secure the global food supply, linking to a Gemma 4 and BOTANIC-1 demo on Hugging Face. The post also offers the BOTANIC-1 download and the accompanying paper.
Google Gemma@googlegemmaOfficialAI score46AIAI lab Living Models pairs Gemma 4 E4B with BOTANIC-1, a genomic language model, to speed up identifying DNA that makes crops climate-resilient. Gemma prepares genomic data and filters candidates, while BOTANIC-1 scores evolutionary impact to pinpoint the causal mutation. In a recent test, the system ranked a target melon yield mutation first out of 2,494 possibilities after an afternoon of computation.
SemiAnalysis@SemiAnalysis_XAI score22AISemiAnalysis says neocloud customers and its own testing point to reliability problems at IREN's Northern BC sites, including multi-day power outages, network upgrades, storage failures, and XID errors. The post adds that many unhappy users rent directly from IREN rather than from resellers of its capacity.
AIMicrosoft's Azure Hardware Systems and Infrastructure team is applying AI agents across its infrastructure lifecycle, starting with cloud supply chain demand planning. Following a "Lean before AI" approach, the company reports that multi-agent workflows cut planning work that took five to seven business days to hours, with approximately 50% less manual effort and cycle time down up to 75% in selected workflows. Microsoft says it is extending the approach to fulfillment, logistics, and fleet operations while keeping human judgment central.
AINVIDIA's cuPhoton targets the computational bottleneck in scientific image pipelines, where data from observatories, telescopes, lasers, and X-ray light sources arrives faster than CPU-bound processing can handle. The source says the bottleneck is usually the whole path from raw sensor data to decision, not one slow kernel. The available text does not give benchmark figures, pricing, or availability details.
OpenRouter@OpenRouterOfficialAI score14AIOpenRouter has released Decision Model Rankings, which show spend share and token share across decision models and task types. Categories include Relevance, Correctness, and Instruction Following. The post says @typesafeai currently leads all categories.
OpenRouter@OpenRouterOfficialAI score5AIOpenRouter has published a rankings page at for users to explore. The post itself provides no further details about what the rankings measure or how they are compiled.
Aravind Srinivas@AravSrinivasXAI score20AIAravind Srinivas posted that "we're living in incredible times," highlighting a fully computer-made animated scene produced from concept art to final edit. The quoted post says the scene was planned in Blender, with reference images generated by Nano Banana and animation and score produced by Seedance 2.5.
Databricks@databricksOfficialAI score22AIDatabricks' Omnigent lets policies check agent actions before they execute, enforcing spending limits and restricting tool use. Policies can also track accumulated risk across a session and require approval once a threshold is reached.
Vercel Developers@vercel_devOfficialAI score18AIVercel says users and their agents can now search trace spans from the Vercel CLI. The example command vercel traces search --status error --since 1h filters spans with error status from the past hour.
Google AI@GoogleAIOfficialAI score18AIGoogle AI linked to a blog post on identifying AI-generated content, pointing readers to its SynthID approach from Google DeepMind. The post itself gives no further details, so the summary cannot specify methods or results.
Google AI@GoogleAIOfficialAI score54AIGoogle is letting anyone check files for SynthID watermarks at synthid.com, covering content from Google and partners including OpenAI, NVIDIA and Kakao. Apple is listed as coming soon. Google says it has watermarked 180 billion images and videos and more than 240,000 years of audio, and the portal handles about 1 million verification requests daily.
AIMistral has released Mistral Large 4, nicknamed Le Chonk, a 1 trillion-parameter model it says can be used and customized by anyone. It is in preview, with a final version due by the end of the month, and is optimized for coding and cyberdefense as well as manufacturing, finance, and electrical engineering tasks. Mistral claims it is the most capable open-weight model developed outside China and says it was trained from scratch rather than through distillation.
Google@GoogleOfficialAI score46AIGoogle says it has watermarked more than 180 billion images and videos, plus 240,000 years of audio, since launching SynthID in 2023. The verification feature is built into Search, the Gemini app, and Chrome, which together handle over 1 million verification requests daily. Google presents the SynthID Detector platform as part of its effort to give users more context about online media.
Google@GoogleOfficialAI score20AIGoogle's SynthID portal at synthid.com lets users upload an image, video, or audio file to check it. The portal scans the file for a SynthID watermark from Google or its partners, which indicates AI generation.
Google@GoogleOfficialAI score45AIGoogle is expanding its SynthID Detector verification portal through partnerships with OpenAI, NVIDIA, Kakao, and soon Apple to improve transparency for AI-generated content. The portal is now available globally in English and lets users check whether media was made with AI.
🚨 AI News | TestingCatalog@testingcatalogXAI score25AIGoogle is reportedly prototyping its own voice agent, internally called "Concierge," alongside preparations for a Gemini 4 release on Antigravity. The post describes the prototype as a very early version and says it is unclear whether it will launch in its current form.
Mistral AI@MistralAIOfficialAI score16AIMistral Large 4 is the top open-source model on Harvey's Legal Agent Benchmark (LAB), according to Mistral AI's post. The post is a short announcement with no further scores or details.
Mistral AI@MistralAIOfficialAI score22AIMistral Large 4 scored highest on AutomationBench, a benchmark of 657 business workflows across simulated workplace apps including Gmail, Google Sheets, Slack, and Salesforce. It finished ahead of Kimi K3, MiMo-V2.6-Pro, and DeepSeek V4 Pro.
Mistral AI@MistralAIOfficialAI score40AIMistral Large 4 is positioned to run general-purpose agents that gather information and produce finished deliverables across complex workflows. The post provides no further details on benchmarks, parameters, pricing, or availability.
DuckDuckGo@DuckDuckGoXAI score34AIDuck.ai now lets users share a chat through a link that recipients can read, continue, or save without needing an account. The post introduces the feature and promises a thread explaining how the sharing design protects privacy.
Google DeepMind@GoogleDeepMindOfficialAI score34AIGoogle DeepMind says SynthID Detector can quickly identify its SynthID watermarks in video, audio, and images, even after edits such as adding filters. The post adds that existing AI identification features in Google Chrome, Google Search, and the Gemini app can also be used.
Google DeepMind@GoogleDeepMindOfficialAI score46AIGoogle DeepMind has made SynthID Detector publicly available, letting anyone check whether online content was generated using Google AI or tools from partners including OpenAI, NVIDIA, and Kakao. Apple is listed as coming soon. The tool is accessible at synthid.com.
AIGoogle says its SynthID detector now supports watermarks from all of its partners and is available to anyone on a new SynthID.com website. Previously, outside users could only check for Google's own watermarks by asking Gemini, and the detector did not flag images from other providers such as ChatGPT.
AIGoogle is making its SynthID Detector available globally in English, letting anyone check whether an image, video, or audio file was made with AI from Google or partners including OpenAI, NVIDIA, Kakao, and soon Apple. The tool joins built-in verification in Search, the Gemini app, and Chrome, which now handle over 1 million requests daily. Google says SynthID has watermarked over 180 billion images and videos and 240,000 years of audio.
Why it matters: The source specifies which vendors' AI media the detector checks, helping readers judge how far the verification covers content they encounter online.
Baidu Inc.@Baidu_IncOfficialAI score2AIBaidu Inc. posted a lighthearted message on X asking whether staring at autumn leaves counts as having plans, noting the post was created with DuMate. The post contains no product details, figures, or feature claims beyond that attribution.
AIOpenAI released 722 mathematical manuscripts in 372 families, produced by an unreleased frontier model, with the average result taking the equivalent of three hours of ChatGPT Pro thinking. The author notes many results are verified in Lean but not all, and suggests mathematics could divide into vast machine-verified work and a compressed human 'effective theory' that people can actually understand.
This story has a top pick“OpenAI releases 719 AI-generated math manuscripts, splitting the mathematics community”
AMD@AMDOfficialAI score20AIHPE's new ProLiant Gen13 servers are powered by AMD 6th Gen EPYC server processors, as announced by AMD. HPE positions the servers for automatic security, intelligent operations, and higher performance for enterprise AI-era workloads.
AIMeta has added new large language model detection to flag seemingly benign ads that covertly direct people to illegal content, and it now checks where ads lead, not just what they show. The company said it actioned 33.2 million pieces of child sexual exploitation content on Facebook and Instagram from January to June 2026, with over 97% found before anyone reported it.
Google Workspace@GoogleWorkspaceOfficialAI score13AIGoogle Workspace says users can prompt Gemini in Google Docs to convert a detailed brief into an on-brand presentation deck. The deck is grounded in work files, emails, and chats, with key milestones and formatting added automatically.
LangChain@LangChainOfficialAI score5AILangChain promotes Interrupt, its agent conference, with a London event and an RSVP link. The post, which was shared by Tavily AI, gives no details on the agenda, speakers, or dates.
PixVerse@PixVerseOfficialAI score8AIPixVerse is promoting a plugin that users can try through a marketing link on its app platform. The post gives no details on the plugin's features, supported models, pricing, or availability beyond the link.
PixVerse@PixVerseOfficialAI score20AIThe PixVerse plugin enables video creation from text, images, or video references inside the chat interface. Users select the plugin, describe a scene or add a reference, then specify model, duration, resolution, and aspect ratio before generating.
AINVIDIA reports that fine-tuned Nemotron models reached gold-medal level at both IOI 2026, scoring 535.4 out of 600, and IMO 2026, scoring 30 out of 42. The IOI run was a live, unofficial, unsupervised benchmark, while IMO proofs were graded by official IMO graders. The post also releases checkpoints, datasets, a new 200-problem benchmark, and inference pipelines on Hugging Face and NeMo-Skills.
Why it matters: The post traces how SFT, RL, and a generate-verify-refine loop turned Nemotron into gold-level specialists for IOI and IMO, with the training and inference details shared.
Teknium 🪽@TekniumXAI score36AIDevelopers are running Hermes on devices such as LilyGO watches, AIPI Lite, desk gadgets, and old Android phones after the Hermes Gadget open SDK and demo were released three days ago. The post credits @NousResearch and says more boards are landing on main through contributor PRs, with the SDK available on GitHub.
Elad Gil@eladgilXAI score50AIElad Gil posted a brief reaction to the moment, without details. The quoted OpenAI post says the company is releasing a range of new mathematical results produced by an internal frontier model, reviewed with the Institute for Advanced Study's Advisory Group on Mathematics and Artificial Intelligence.
👩💻 Paige Bailey@DynamicWebPaigeXAI score10AIPaige Bailey took an early morning Waymo ride on the highway down to South Bay for recordings and customer interviews. The post is a brief personal update about the trip and does not describe any product feature or announcement.
