AI Pulse · Releases (daily)

Releases · Sat 3 Oct 2026

Sun, 04 Oct 2026 00:00:00 +0000

"Muse Gadgets" turns AI hardware into an open-source DIY project
Meta announced Muse Gadgets, an open-source project that lets hobbyists build their own AI hardware using ESP32 boards and connect it to Meta's AI agent Muse. (The Decoder)
Prime Intellect Launches Prime Inference: Serverless and Reserved Serving for Frontier Open Models
Prime Intellect has launched Prime Inference, an OpenAI-compatible platform for serving frontier open models on NVIDIA Blackwell. (MarkTechPost)
Open-source "BootLoops" harness supports AI models in performing precise scientific calculations
Harvard physicist Matthew Schwartz used the open-source tool BootLoops and Claude to produce 36 manuscripts across 18 fields in three months, from particle physics to linguistics. (The Decoder)

Releases · Fri 2 Oct 2026

Sat, 03 Oct 2026 00:00:00 +0000

AI music generator Suno can now create spoken audio with matching background music
Suno is adding a feature called "Speech" to its AI music generator that creates spoken text with matching background music in a single audio track. (The Decoder)
Add secure Web Search to Claude Desktop with Amazon Bedrock AgentCore
Claude Desktop on Amazon Bedrock is limited to the model's knowledge cutoff without web search. (AWS Machine Learning Blog)
Open-sourcing AstaBrief, the fast report-generation model in Asta
Language models can already help researchers search the literature, synthesize evidence, and work through complex questions. (Hugging Face Blog)
A model guide for the GPT-6 family
Learn how startups can choose GPT-6 models, tune reasoning effort, improve prompts and skills, coordinate tools, and prepare workflows for production. (OpenAI News)
Introducing Web Search API via AI Gateway
Cloudflare AI Gateway now supports native web search API integration in partnership with Ceramic.ai, Exa, and Linkup. (Cloudflare Blog)
NVIDIA DGX Spark 64GB Gives Developers More Ways to Build and Scale Local AI
Local AI is becoming more useful by the token. As AI agents move from experiments into everyday development, increasingly capable open models are shrinking to fit on more devices, giving builders more to run locally. (NVIDIA Blog)
Toward provably private learning from federated data
We announce a new Federated Learning system that provides externally verifiable privacy guarantees while shifting computation to the server to improve training speed, accuracy, and device coverage. (Google Research Blog)

Releases · Thu 1 Oct 2026

Fri, 02 Oct 2026 00:00:00 +0000

Google thinks SpaceX’s Starship has to launch 1,800 times before space data centers get off the ground
Google launched its first advanced chip into orbit to pave the way for space data centers. (TechCrunch AI)
AWS debuts Strands Decider 2B, a first lightweight decision model for accelerate agentic workflows
Amazon Web Services Inc.’s Strands Labs team has been playing around with an emerging class of lightweight artificial intelligence systems known as “decision models,” and it’s now making the fruits of that work available to the open-source… (SiliconANGLE AI)
ServiceNow launches conversational interface to its service desk
ServiceNow Inc. is enhancing the artificial intelligence features in its service desk offering with Flow, a feature that handles employee requests via a natural language interface with extensive automation. (SiliconANGLE AI)
Ideogram says its new model can edit part of an image without messing up the rest
Ideogram's new model Ideogram 4.5 promises to edit only the areas you want while keeping the rest of the image intact. (The Decoder)
ChatGPT can now virtually try on clothes for you
OpenAI is rolling out new shopping features for ChatGPT that let users virtually try on clothing and accessories using their own photos and save products they like to a Favorites library. (TechCrunch AI)
Announcing Ranveer Singh as Brand Ambassador for Ray-Ban and Ray-Ban Meta in India along with Exciting New Updates to our AI Glasses
Ranveer Singh becomes the first Brand Ambassador for Ray-Ban and Ray-Ban Meta in India. (Meta Newsroom)
Introducing Clef: our open-source decision models, and new RL fine-tuning platform
We are introducing Clef and Clef-flash, open-source decision models hosted on Workers AI for high-speed classification and agentic workflows. (Cloudflare Blog)
Introducing Olmo-core 3: Open, scalable training infrastructure for large MoEs
Today we’re releasing Olmo-core 3, a significant upgrade to our framework for developing large language models featuring a redesigned open mixture-of-experts (MoE) training system. (Hugging Face Blog)
One year later: Sovereign AI and the fight for choice
AI sovereignty is not a zero-sum game, but many governments now believe it is. Cloudflare's answer: more local open-source models, model-agnostic security tools, and a commitment to giving nations genuine choice. (Cloudflare Blog)
AI Search is now generally available
It embeds image pixels directly for visual search, runs optical character recognition on scanned PDFs, accepts files up to 10 MiB, and works with any chat model. (Cloudflare Blog)
Barracuda launches integrated e-mail protection for AI-era protection
Barracuda IQ uses multi-model AI to analyse intent and identity behaviour and multi-modal content across over 100 global threat feeds. (ITWeb)
Bilibili Open-Sources Index-Translate, a Qwen3.5-Based Translation Model Family for 150 Languages
Bilibili's Index LLM team released Index-Translate under Apache-2.0: 2B, 9B and 35B-A3B preview text models for 150 languages, plus dubbing, syllable-control and long-document variants. (Pandaily)
Perplexity Releases pplx-embed-v2-context-9b-preview: A Contextual Embedding Model That Retrieves Answers and Their Supporting Evidence
Perplexity Research and turbopuffer have released pplx-embed-v2-context-9b-preview, a contextual embedding model for RAG pipelines. (MarkTechPost)
Barclays scales Claude to upgrade operations and improve client experience
Barclays, the British universal bank, is expanding its strategic collaboration with Anthropic to integrate secure, enterprise-grade AI systems across its global operations. (Anthropic News)

Releases · Wed 30 Sep 2026

Thu, 01 Oct 2026 00:00:00 +0000

Introducing ai_decide: make fast decisions on your governed data
Today, we’re launching a new Databricks AI Function ai_decide that makes fast decisions… (Databricks Blog)
DoorDash launches an AI agent you can text to order food
By launching an AI agent for food ordering, DoorDash is looking to gain an edge over rivals Uber Eats and Grubhub. (TechCrunch AI)
Introducing SynthID Bio
Proof of concept for watermarking AI-generated proteins while preserving biological function. (Google DeepMind Blog)
Airbnb adds AI search, more social features
Airbnb is also launching new services such as meal delivery and laundry in select locations. (TechCrunch AI)
Monetization Gateway beta: charge AI agents for consumption with HTTP 402
Cloudflare’s AI Gateway, Ceramic.ai, Stocktwits, and more are using the Cloudflare Monetization Gateway today to charge agents for access to tokens, APIs, and MCP tools. (Cloudflare Blog)
Pay Per Use: when AI uses your work, you should get paid
Pay Per Use is now in beta. AI companies report when they use publishers’ content, and Cloudflare handles billing, payouts, and reporting, so our customers are paid according to use. (Cloudflare Blog)
Identify AI model overuse with User Insights
AI Gateway User Insights now adds task, model, turn, and user categories to help teams understand AI adoption and make better model decisions. (Cloudflare Blog)
Cloudflare Containers, rebuilt to scale agent sandboxes
Cloudflare Containers now start 6x faster, let your agent choose each sandbox's image and instance type at runtime, and support filesystem snapshots in public beta, all controlled from a Durable Object. (Cloudflare Blog)
CNA to relaunch Spanish-language Focus Taiwan with AI assistance
Taipei, Sept. 30 (CNA) Taiwan's Central News Agency (CNA) announced Wednesday that it will relaunch its Spanish-language Focus Taiwan news service Thursday, using artificial intelligence (AI) to help produce and edit stories. (Focus Taiwan)
Perplexity Introduces Photon: A Rust-Based Retrieval Engine That Cuts p99 Latency From 800 ms to 65 ms
Perplexity has released Photon, an in-house retrieval and ranking engine written in Rust. It replaces an open-source engine Perplexity had forked for its AI-native search stack. (MarkTechPost)
Anthropic says Zhipu's open-weight GLM-5.3 nearly matches Claude Mythos Preview at building exploits
Zhipu's open-weight model GLM-5.3 writes cyber exploits nearly as well as Claude Mythos Preview, according to Anthropic. (The Decoder)
Cohere Embed 5: Frontier Embedding Models for Enterprise
Cohere's Embed 5 delivers state-of-the-art retrieval across multimodal, multilingual, and financial documents - now available in Pro and Fast tiers. (Cohere)
Simate Debuts Simate-beta Physical AI Fast System, Says It Tops RoboDojo Benchmark
Three-month-old Simate released Simate-beta, a fast-reaction physical AI model it says ranks first on RoboDojo with a 33.95 average score, developed with an agent-driven research pipeline. (Pandaily)
IQuest Research Open-Sources IQuest-Q1, a 320B MoE Model for Agentic Coding With 15B Active Parameters
IQuest Research released open weights for IQuest-Q1, a 320B sparse MoE with 15B active parameters built for command-line coding agents, with 512K context and 84.5 on CyberGym. (Pandaily)
DeepSeek Open-Sources Ascend Versions of TileLang, DeepGEMM and DeepEP as Huawei Details SuperPoD Flex
DeepSeek released Ascend versions of TileLang, DeepGEMM, DeepEP, TileKernels, FlashMLA and DeepSelect, while Huawei detailed a jointly defined 128-card SuperPoD Flex design. (Pandaily)
ChatGPT maker wants to be the App Store for AI as safety concerns grow
OpenAI on Tuesday launched a less expensive artificial intelligence (AI) model and tools that run continuously, as it aims to reposition ChatGPT as an app store for AI. (Tech Xplore)
DeepSeek Harness v0.2 Preview Adds Desktop Installers, Plugin Manager and Scheduled Automations
DeepSeek released Harness v0.2 preview with official macOS and Windows installers, a plugin manager that installs npm packages without a command line, file previews and scheduled automations. (Pandaily)
Amazon Bedrock expands Claude model availability to in-country inferencing in India
Anthropic's Claude Opus 5, Claude Sonnet 5, and Claude Haiku 4.5 are now available in India through Amazon Bedrock geographic cross-Region inference. (AWS Machine Learning Blog)

Releases · Tue 29 Sep 2026

Wed, 30 Sep 2026 00:00:00 +0000

NetApp and Oracle to Launch Fully Managed Cloud Storage Service for Demanding AI and Enterprise Workloads
NetApp® (NASDAQ: NTAP), the Intelligent Data Infrastructure company, and Oracle today announced a new fully managed storage service that brings enterprise-grade NetApp storage natively to Oracle Cloud Infrastructure (OCI). (ITWeb)
Equals Money lets customers’ AI tools read data but not move money
Opening a Model Context Protocol server, or MCP server, to customers’ AI tools is becoming standard for fintech companies, but in payments a rogue agent does not just leak data — it moves money. (SiliconANGLE AI)
The internet is convinced Elon Musk’s xAI trolled OpenAI’s ‘Dots’ launch
Before OpenAI launched its new AI agent, Dots, on Tuesday, Elon Musk's xAI had already acquired the domain name "dot.com," which now redirects to the Grok chatbot download page. (TechCrunch AI)
OpenAI unveils low-cost AI model, a day after shelving Astra upgrade
OpenAI launched a new midrange artificial intelligence model Tuesday at one-fifth the cost of its flagship offering, one day after scrapping an upgraded version deemed unreliable. (Tech Xplore)
Liquid AI Releases d1: A Decision Model That Returns Calibrated Probabilities With Zero Output Tokens
Liquid AI has released d1, a decision model built for structured choices instead of text generation. You give it context and a set of typed questions. (MarkTechPost)
OpenAI’s latest features take direct aim at the app store model
OpenAI is building out the pieces of an alternative to the traditional app store model, turning ChatGPT into a place where software can be discovered and used by people and AI agents alike. (TechCrunch AI)
Introducing GPT-6.1 Sol
Meet GPT-6.1 Sol: near-Astra intelligence for coding, computer use, and professional work at one-fifth of Astra’s standard API input and output token prices. (OpenAI News)
Introducing dots
Dots by OpenAI are a proactive assistant that can keep working across complex projects and everyday tasks. Learn how dots help you stay in control while work moves forward. (OpenAI News)
OpenAI takes on Microsoft with the launch of what feels a whole lot like ChatGPT’s own office suite
OpenAI's newly announced suite of office features puts it into mor direct competition with more traditional software companies. (TechCrunch AI)
OpenAI expands ChatGPT’s plugins with app-like interfaces and automations
OpenAI is expanding ChatGPT plugins with dedicated sidebar homes, interactive panels, file viewers, improved discovery, and support for automations. (TechCrunch AI)
Prompt engineering by Quick component: Patterns and pitfalls
Part 2 of our Amazon Quick prompt engineering series goes component by component. (AWS Machine Learning Blog)
ElevenLabs' new v4 speech model makes AI voices more expressive and consistent
Elevenlabs' new speech model, Eleven v4, follows cues for laughter and whispering more accurately and keeps voices consistent across long productions like audiobooks. (The Decoder)
NVIDIA Kumo Tabular Sets a New Accuracy-Efficiency Frontier for Tabular Prediction
NVIDIA Kumo Tabular, part of the NVIDIA Kumo Structured model collection, is an open foundation model for tabular data now available on Hugging Face. (Hugging Face Blog)
Introducing Quine: An AI research system designed for the complexity of biology
Biology doesn't operate in silos, and neither should the AI representation of it. Quine is an early-stage research effort to create a multimodal world model of biology. (Microsoft Research)
Oracle expands AI agent features with Fusion Claw
Oracle Corp. today is introducing an agentic software runtime that lets its Fusion applications carry out complex business tasks, from reconciling accounting entries to revising staffing plans when conditions change. (SiliconANGLE AI)
The open-source AI platforms vying to become China’s Hugging Face
AI model platforms ModelScope and MoArk are competing to serve Chinese-speaking developers behind the Great Firewall. (Rest of World)
Kling AI Previews Kling 4.0 With 30-Second Native Clips and Multi-Reference Control
Kuaishou's Kling AI announced Kling 4.0 with single-pass clips of up to 30 seconds, up to 15 multimodal references and 10 keyframes; Kling 4.0 Flash opened to Ultra Yearly subscribers on September 28, with the full model due in October. (Pandaily)
Alibaba's 27B XekRung Model Tops CyberGym Model Leaderboard at 88.9%
Alibaba Security AGI Lab's XekRung-1.5-27B-Preview, fine-tuned from Qwen3.8-27B, reached an 88.9% success rate on UC Berkeley's CyberGym Level 1 and now ranks first on the benchmark's model leaderboard. (Pandaily)
Google Research Open-Sources RRSI: AI Agents That Improve Their Own Harness Without Overfitting
Google Cloud AI Research has open-sourced RRSI, a framework that lets LLM agents rewrite their own prompts, tools and memory while model weights stay frozen. (MarkTechPost)
Alibaba Qwen Releases Qwen-Audio-3.1-Realtime: A Full-Duplex Voice Model Trained to Think, Act, and Decide When to Speak
Alibaba's Qwen team released Qwen-Audio-3.1-Realtime, a full-duplex voice model trained to reason, call tools and decide when to speak. (MarkTechPost)
Anthropic Releases Claude Sonnet 5.5: 70.6% on Terminal-Bench 4.0 at the Same $2/$10 Price
Anthropic has released Claude Sonnet 5.5, the second model in its Claude 5.5 family. It scores 70.6% on Terminal-Bench 4.0 and lands within 2 points of Opus 5.5 on GDPval-AA. (MarkTechPost)
Manus 2.0 lets users edit videos, host multiplayer games, and run agents remotely from their phone
Manus is turning its AI agent into a platform with version 2.0, letting users edit videos, host multiplayer games, and run personal agents with their own phone numbers. (The Decoder)
GPT-6.1 Astra is too deceptive for release, marking OpenAI's most dramatic safety intervention yet
OpenAI has halted the release of GPT-6.1 Astra after internal tests found it acted without permission, misled users, and accessed external services despite safety risks. (The Decoder)
Delta Intelligence Ships Delta-0 Native Humanoid Foundation Model
On September 28, 2026 Delta Intelligence released Delta-0, a brain-cerebellum-force humanoid foundation model that demoed 32 planned steps and nine household tasks in about 210 seconds of continuous loco-manipulation. (Pandaily)

Releases · Mon 28 Sep 2026

Tue, 29 Sep 2026 00:00:00 +0000

Grok 4.7 is now available on Amazon Bedrock
xAI's Grok 4.7 is now available on Amazon Bedrock: a frontier model for coding, long-running agents, and knowledge work. (AWS Machine Learning Blog)
introducing PipSqueak 3, ShortSqueak, and (c.ai) lite
Two new free Chat Styles are rolling out to everyone on (c.ai): PipSqueak 3 , a big step up for general roleplay, and ShortSqueak , a much requested style for quick, casual chats. (Character.AI)
Basis completes a tax workbook 2x faster with GPT-6 Astra
GPT-6 Astra completed a 50-tab tax workbook twice as fast as GPT-5.6 Sol, and its stronger understanding of user intent gives Basis more confidence in real-world use. (OpenAI News)
Introducing Claude Sonnet 5.5
A clear upgrade over Sonnet 5 that runs 30% faster and costs up to 30% less for most work. (Anthropic News)
NVIDIA Launches Open Agent Safety Platform: OpenShell Sandboxes Agents on Vera CPUs While Sentry on BlueField-4 Quarantines Them in Milliseconds
NVIDIA has launched the Open Agent Safety Platform, an open reference design that enforces AI agent safety outside the agent itself. (MarkTechPost)
Incognito Software Systems Launches NEXA, an AI-Enabled Solution for Automated Broadband Operations
Incognito Software Systems, a leading global provider of service orchestration, device management, and network intelligence solutions, today introduced NEXA… (ITWeb)
Fireworks AI Releases Ember-1: A Post-Trained Kimi K3 That Uses About 40% Fewer Tokens
Fireworks AI has released Ember-1, a post-trained Kimi K3 that learns to produce shorter reasoning traces instead of lowering reasoning effort. (MarkTechPost)
Google Research Introduces an AI Video Co-Director: 4 Agentic Frameworks for Coherent, Minutes-Long Video Generation
Google Research has introduced an AI video co-director for long-form video generation. The suite of 4 agentic frameworks turns short clips into coherent, minutes-long stories. (MarkTechPost)
Next.js applications, powered by Vite: introducing Vinext 1.0
Vinext 1.0 graduates from an AI experiment to a production-ready framework, letting developers run Next.js apps on Vite. (Cloudflare Blog)
Rubrik brings agentic AI to cyber recovery with launch of Rubrik MCP
Rubrik Model Context Protocol gives an organisation’s AI agents a secure, programmable path to Rubrik’s data, identity and application intelligence. (ITWeb)
Untapped AI files patent for the handshake that lets AI agents prove who they are before they do business; will offer it free to SA companies
Built for the age of AI assistants calling AI businesses on behalf of consumers, the method lets machines confirm who they represent before a single sensitive detail is shared. (ITWeb)
China’s first AI-produced theatrical film sets October 23 release
AI-generated video is no longer anything new. But a 100-minute theatrical feature film that uses AI throughout the production process and has secured an official screening license shows how far AI filmmaking has come. (TechNode)
DeepSeek Harness Desktop Preview: Electron Packages on Win and Mac
DeepSeek Harness desktop preview Electron packages for Windows and macOS appeared on download.deepseek.com around September 25, 2026, framed as RC/nightly builds rather than a GA launch. (Pandaily)
StepFun Step 5 Preview: ~600B MoE Long-Horizon Agent; Open Weights Due Oct 15
StepFun launched Step 5 Preview on September 20, 2026: a ~600B sparse MoE (~27B active) with 1M context for long-horizon agents, API live now and BF16 open weights planned for October 15. (Pandaily)
NaiveAI Open-Weights Naive-N0.5-Flash: 309B MoE, Hybrid SWA–DSA 1M Context, MIT
NaiveAI released Naive-N0.5-Flash as MIT open weights: a 309B MoE with 15.5B active, hybrid SWA–DSA native 1M context for coding and AI R&D, built on MiMo-V2.5 without full-attention layers. (Pandaily)
Meituan Ships LongCat-2.5-Preview: 1.6T MoE Multimodal Agent Model With 1M Context
Meituan launched LongCat-2.5-Preview on September 25, 2026: a 1.6T MoE (~48B active) with native 1M-token multimodal context aimed at long-horizon software agents, without publishing benchmarks this round. (Pandaily)
Holo4: powering generalist computer-use agents
Holo4 is our new series of agentic models. It comes in two sizes: 27B dense and 35B-A3B Mixture of Experts. (Hugging Face Blog)

Releases · Sun 27 Sep 2026

Mon, 28 Sep 2026 00:00:00 +0000

SAIL: Scaling In-Context Imitation Learning
Introducing “Scaling In-Context Imitation Learning” (SAIL) to be presented at IROS2026. This work is a collaboration between Sakana AI and the University of Tokyo. (Sakana AI)
Nvidia drops a free 100M-parameter model that identifies up to eight speakers in real time
Nvidia released Nemotron 3 Diarization, an AI model that identifies which speaker is talking at any given moment in a conversation. (The Decoder)

Releases · Sat 26 Sep 2026

Sun, 27 Sep 2026 00:00:00 +0000

Sarvam AI Releases Saaras V4: A Speech-to-Text Model for All 22 Indian Languages and Global English
Sarvam AI's Saaras V4 is a speech-to-text model covering all 22 Indian languages plus global English. It pairs an audio encoder with a 3B hybrid state-space decoder. (MarkTechPost)
Supersonic Labs Releases Julia 1: A 144.3M-Parameter Open Decision Model That Runs on a CPU
Supersonic Labs has released Julia 1, a 144.3M-parameter decision model built on mmBERT-small. It takes context, a question, and 2 to 20 options, then returns one choice with probabilities. (MarkTechPost)
Exa Launches Agent Ultra: A Subagent Swarm Deep Research API Built for Exhaustive List Building
Exa has released Agent Ultra, the highest effort mode of its Exa Agent API. It coordinates subagents across thousands of sources for list building and entity enrichment. (MarkTechPost)
OpenAI's GPT-6 Astra can now tell you exactly where you screwed up your IKEA shelf
OpenAI's GPT-6 Astra can look at a photo and tell whether an IKEA furniture piece was assembled incorrectly, hitting an 80 percent accuracy rate. (The Decoder)

Releases · Fri 25 Sep 2026

Sat, 26 Sep 2026 00:00:00 +0000

Aikido Security Releases Altar-1: An Open-Weight Security Model Pruned From GLM-5.3 to 328 GB
Aikido Security has released Altar-1, its first open-weight security model. It is a compressed version of Z.AI’s GLM-5.3, built to run inside infrastructure the customer controls. (MarkTechPost)
Fastino Releases GLiNER2.5-Decide: A 340M Open-Weight Decision Model That Runs on CPU
Fastino Labs has released GLiNER2.5-Decide, a 340M-parameter open-weight decision model. It takes text and a schema of typed questions and returns structured answers. (MarkTechPost)
Microsoft overhauls Copilot with new coding, document editing features
Microsoft Corp. today launched a new version of Copilot, the artificial intelligence assistant included in its Microsoft 365 productivity suite. (SiliconANGLE AI)
Warp adds programmable agents to its AI-native HR platform
Warp today launched an artificial intelligence agent designed to automate routine human resources work, extending its AI-native employee management platform with customizable workflows that can act without continual human prompting. (SiliconANGLE AI)
Agents can now set up your website’s security with Turnstile Spin
Misconfiguring Turnstile by skipping backend validation leaves sites exposed to bots. Turnstile Spin fixes incomplete setups by using your preferred AI coding agent to wire up server-side verification. (Cloudflare Blog)

Releases · Thu 24 Sep 2026

Fri, 25 Sep 2026 00:00:00 +0000

Contrastive-LM Releases CLM-8B: An Open System One Model That Scores Agent Actions Up to 9× Faster Than Jev
Contrastive-LM has released CLM-8B, an open System One model that scores candidate actions against a state instead of generating text. (MarkTechPost)
BottleCap AI Releases ThinkingCap-Qwen3.8-27B: 37.2% Fewer Thinking Tokens at a 0.86pp Accuracy Cost
BottleCap AI has released ThinkingCap-Qwen3.8-27B, a fine-tune of Qwen3.8-27B that spends 37.2% fewer thinking tokens across 12 benchmarks. (MarkTechPost)
Amazon’s new plugin lets sellers manage their entire e-commerce business with AI
Amazon.com Inc. said today it’s trying to make life easier for independent sellers using its ecommerce platform with new tools including a plugin that allows them to run their businesses via artificial intelligence assistants. (SiliconANGLE AI)
Automating coherent long-form video generation
We introduce a unified multi-agent framework that autonomously generates temporally consistent, long-form video narratives, overcoming the identity drift and cascading failures of current linear AI pipelines. (Google Research Blog)
Black Forest Labs launches FLUX 3 Action, an open robotics AI model
Black Forest Labs is entering robotics with FLUX 3 Action. The open-world-action model uses camera feeds to predict what action a robot should take next. (The Decoder)
Google Photos ‘Clueless’-inspired virtual closet is now available on Android and iOS
The AI-powered feature builds a virtual wardrobe from your photos, and is now broadly available after first rolling out to Android users in June. (TechCrunch AI)
Ando wants to take on Slack with a team messaging app that lets humans and agents work together
The app gives agents their own identities and inboxes and lets them partake in conversations as naturally as people can. (TechCrunch AI)
Accelerating vision-language models with LFM2.5-VL-DSpark
Today, we release an experimental DSpark draft model for our vision-language model (VLM) LFM2.5-VL-3B. (Hugging Face Blog)

Releases · Wed 23 Sep 2026

Thu, 24 Sep 2026 00:00:00 +0000

Anthropic, OpenAI release cheaper AI even as safety fears grow
Anthropic and OpenAI on Tuesday released new, cheaper artificial intelligence models within hours of each other, days after the heads of both companies called for a slowdown in AI development. (Tech Xplore)
NVIDIA Releases Nemotron 3 Diarization: A 100M-Parameter Open-Weight Model That Tracks 8 Speakers in Real Time
NVIDIA has released Nemotron 3 Diarization, an open-weight speaker diarization model on Hugging Face. It answers one question about any conversation: who spoke when. (MarkTechPost)
Nokia Open-Sources AnyJev: A Training-Free Layer That Turns Any Open LLM Into a Calibrated Decision Model
Nokia’s applied research team has open-sourced AnyJev, a Python library that turns an open LLM into a decision model. (MarkTechPost)
Kyutai Releases Voice of Reason: A Speech-Native Model that Solves Spoken Math with Reinforcement Learning
Kyutai has released Voice of Reason, 2 open-weight speech-to-speech models built on GLM-4-Voice-9B. Supervised fine-tuning and reinforcement learning lift spoken GSM8K accuracy from 27.3% to 77.1%. (MarkTechPost)
ZeroDrift launches three models for real-time AI compliance checks
ZeroDrift Inc., a startup that automates compliance for artificial intelligence communications, today launched Anchor 3.0, a family of small language models designed to check messages generated by AI agents before they’re sent. (SiliconANGLE AI)
Alibaba launches Qwen Audio 3.1 with new models and slashes AI audio prices by up to 95 percent
Alibaba's AI team Qwen has released Qwen-Audio-3.1, a lineup of five models for speech recognition (ASR), text-to-speech (TTS), and real-time interaction. (The Decoder)
YouTube Music gets more conversational with new AI features
Built directly into the YouTube Music app, Ask Music lets users describe what they want to hear in everyday language rather than searching for individual songs or artists. (TechCrunch AI)
YouTube releases new AI features for creators within its Studio app
YouTube is adding new features to generate ideas and monitor the performance of thumbnails. (TechCrunch AI)
Use open weight models as your AI coding agent with Amazon Bedrock
Pair OpenCode, an open-source terminal-native AI coding agent, with open weight models on Amazon Bedrock to get a secure, flexible, pay-per-use coding assistant. (AWS Machine Learning Blog)
Advancing Private AI Compute with secure, server-side memory
Introducing private, server-side memory to Private AI Compute for personal AI. (Google DeepMind Blog)
Two years of OpenAI Academy
Marking two years of OpenAI Academy and bringing AI skills to even more communities. (OpenAI News)
OpenAI extends cyber access to Ukraine for civilian defense
OpenAI is extending access to its Daybreak program to the Government of Ukraine to support the cyber defense of civilian infrastructure. (OpenAI News)
Introducing MentalHealthBench
MentalHealthBench is an expert-informed benchmark for evaluating helpful and safe AI responses across realistic mental health conversations. (OpenAI News)
ChatGPT Ads expands to Southeast Asia and Taiwan
ChatGPT Ads is expanding to Southeast Asia and Taiwan, giving eligible businesses new ways to reach people across more than 60 countries. (OpenAI News)

Releases · Tue 22 Sep 2026

Wed, 23 Sep 2026 00:00:00 +0000

Introducing Claude Opus 5.5
Opus 5.5 performs at the level of Claude Fable 5.1 on most work and costs 40% less to run than Opus 5. (Anthropic News)
China's Alibaba unveils new powerful chip and ambitious AI model plans
China's Alibaba unveiled new artificial intelligence chip technologies and plans for more powerful models on Tuesday, including what it said was China's most powerful AI chip… (Tech Xplore)
SpaceXAI Releases Grok 4.7: A Larger Base Model at the Same $2/$6 Price as Grok 4.6
SpaceXAI has released Grok 4.7, its new flagship model for coding, agentic tasks, and knowledge work. Grok 4.7 is built on a larger base model and a longer reinforcement learning run. (MarkTechPost)
NVIDIA Introduces SoL-Pi: Auto-Research Loops That Cut Coding Agent Token Traffic by Up to 49%
NVIDIA researchers have released SoL-Pi, 4 harness mechanisms for the open-source Pi coding agent, discovered by an AI running auto-research loops across 535 environments. (MarkTechPost)
Xiaomi introduces Mimo-V2.6 series open-source AI model family
Global technology company Xiaomi Corp. today announced the release and open-sourcing of its MiMo-V2.6 series generative artificial intelligence models today, which includes two natively omnimodal models bringing a balance of intelligence… (SiliconANGLE AI)
SpaceX launches Grok 4.7 with long-horizon processing, safety upgrades
SpaceX Corp. today introduced Grok 4.7, its most capable large language model to date. The Grok algorithm series was originally created by an Elon Musk-founded startup called xAI Corp. Musk merged the company with xAI Inc. last year. (SiliconANGLE AI)
Qualcomm launches two new smartphone chips with emphasis on AI
Qualcomm said that its new top chip can run 30B mixture-of-expert model locally. (TechCrunch AI)
Anthropic releases Opus 5.5 with lower prices and Fable-level performance
Anthropic called it "the strongest-performing model we've tested to date." (TechCrunch AI)
Genie One MCP: Give any AI Agent the Right Business Context
Business leaders often have access to plenty of data, but still can’t get a reliable… (Databricks Blog)
The Genie One MCP is now Generally Available
AI coworkers and coding agents are spreading fast across organizations, and each… (Databricks Blog)
NVIDIA Isaac ROS 5.0 Advances Agentic, Open Source Robotics Development
To build and deploy sophisticated robotics applications that can perceive, reason and act in dynamic environments, developers need new physical AI models and tools. (NVIDIA Blog)
Better prompt caching for GPT-6
Learn how GPT-6 improves prompt caching with higher cache hit rates, new diagnostics, explicit breakpoints, and controls that reduce latency and costs. (OpenAI News)
Introducing GPT-6 Sol and Luna
Meet GPT-6 Sol and Luna, two models that bring frontier intelligence to everyday work with different balances of capability and cost. (OpenAI News)

Releases · Mon 21 Sep 2026

Tue, 22 Sep 2026 00:00:00 +0000

StepFun Launches Step 5 Preview: A 600B-Total, 27B-Active MoE Model With 1M Context for Long-Horizon Agentic Work
StepFun has released Step 5 Preview, a sparse Mixture-of-Experts model with 600B total parameters and 27B active per token. (MarkTechPost)
AWS debuts Strands Harness, an open-source AI agent that can be deployed in any environment
Amazon Web Services Inc. says it’s trying to help developers solve the problem of scaling artificial intelligence agents to cloud environments with the launch of Strands Harness… (SiliconANGLE AI)
xAI launches Grok 4.7 at bargain prices, but benchmarks reveal a wide gap to Claude and GPT-6
xAI has released Grok 4.7, its most capable model yet. But on the Artificial Analysis Intelligence Index, it scores just 46 points, landing mid-pack and well behind Claude Fable 5.1 and GPT-6 at 53 each. (The Decoder)
ByteDance launches Dramagic, a full-pipeline AI platform for producing short dramas from script to screen
Bytedance launched Dramagic, an AI platform that handles the entire short drama production pipeline, from script to video preview. (The Decoder)
Python Workers are now generally available
Python Workers allow developers to run Python web frameworks and AI orchestration libraries natively in the Cloudflare Workers runtime. (Cloudflare Blog)
NVIDIA Launches DSX Ready to Qualify Power and Cooling Products for AI Factories
Every AI factory needs power and cooling that fit its computing architecture. As AI infrastructure expands, power, cooling, water, site and grid constraints are shaping what builders can deploy. (NVIDIA Blog)
tokenizers v1: encode, decode and scaling, measured
The tokenizer has not historically been the bottleneck within ML workflows. Compute-wise, tokenization is light compared to the heavy modeling happening in the rest of the pipeline. (Hugging Face Blog)
Higgsfield AI ships new video features in a day with GPT-6 Astra
With GPT-6 Astra, Higgsfield AI makes video ad creation easier for small businesses and brings new creative tools to market faster. (OpenAI News)

Releases · Sun 20 Sep 2026

Mon, 21 Sep 2026 00:00:00 +0000

Alibaba Qwen Team Releases Qwen3.8-LiveTranslate: A Real-Time Interpretation Model That Cuts Average Lag to 2.3 Seconds Across 60 Languages
Qwen has released Qwen3.8-LiveTranslate, a real-time simultaneous interpretation model built on a new Interleave architecture. (MarkTechPost)
Alibaba's open-weight Qwen-Image-2.1 claims to beat closed models in image generation with just 7 billion parameters
Alibaba's Qwen team has released Qwen-Image-2.1, an open-weight model that generates and edits images on powerful consumer GPUs, with support for transparency and up to ten reference images at once. (The Decoder)