All entries

A thing to know about the AI business is labs that have frontier models can release half-built products and they work surprisingly well because the AI can just figure stuff out and improvise. Its like

After a brief period where OpenAI seemed to be unifying work around the ChatGPT app, between Dot and Spaces and Pages and local/cloud ChatGPT Work and Scheduled Tasks in the Cloud and Scheduled Tasks

Agents write application code, but you still get paged at 2am when it breaks. We wanted to know whether AI could handle that part of the job too. Introducing Incident Arena: a benchmark that puts codi

Announcing our first open-weights world-action model, Praxis-1. It turns out that, unsurprisingly, you can get predictably better robotics policy performance by scaling third-person video. It’s been f

**Ollama Blog – “Ollama Now Supports JEW‑Style Decision Models” (Sep 30 2026).** The post announces that Ollama now includes native support for JEW‑style decision models, expanding its generative AI c

Breakthroughs on every frontier. I can't wait to share more on what Praxis-1 will unlock across industries. Media Introducing Praxis-1, an open-weight World Action Model that turns our video pre-train

Cognition is the first customer running @NVIDIA Vera Rubin, powered by @CoreWeave! Hopper in 2022. Blackwell in 2024. Now Vera Rubin: On SWE-2 inference, the new chips deliver about 4.8x more token th

Data centers on the ground are limited by permits and power. SpaceX's answer is Starmind, orbital AI data centers with about 72 Nvidia chips and up to 175 kW per satellite. That's about 5,700 satellit

🪽DFlash draft models are available for Ornith-1.5 9B, 35B-A3B, and 397B. DFlash and Ornith 1.5 make a dynamic duo: DFlash performs as the draft model, accelerating inference by drafting multiple token

GLM 5.3 Flash is now available for training on the Serverless Training API - accessible to all, with both vision and text support. This model performs great on our benchmarks for agentic coding, docum

Grok 4.7 rank 1 on AA cyber index Grok 4.7 xHigh ranks #1 in Artificial Analysis Cyber Index It's a powerful frontier model performing at the highest level in enterprise cyber defense Outperforming Fa

Half a million downloads in a month. Today, our open source family takes another step forward. Thank you for the incredible support behind our first-generation models. We’re excited to introduce TwIL-

Hugging Face (HF) provides a platform designed to host large‑scale robotics datasets. An example implementation can be seen in the space at `https://huggingface.co/spaces/imstevenpmwork/LeRobot_and…`,

How many people show up on a Tuesday night to talk about document processing for AI agents? 🤔 In New York, enough to fill the room and start a waitlist. In San Francisco, almost 600. We were in both c

I got thousands of DMs and it looks like @bot can't automatically analyze DMs (when @elonmusk?) so it will take me a few days to go through all of them, sorry! If you don't get an answer, don't take i

i kinda wish dots was a standalone app and not part of chatgpt. as it stands, it feels like a product feature (e.g. GPTs) and not a big, exciting new release. i might understand why they did it though

The user expresses enthusiasm for the decision‑modeling approach introduced by @typesafeai’s JEV tool, emphasizing a preference for local deployment to protect sensitive data. After learning that @oll

In light of the latest Anthropic's latest not-particularly-oblique attempt to build a case against open weights, I thought I would write down how I currently think about open weight AI models. (1) Ope

Independent numbers from @ValsAI. Fewer tokens per turn means lower cost and faster agent loops on every call. Try Ember-1 on Fireworks Serverless: https://fireworks.ai/models/fireworks/ember-1 Firewo

Introducing MCP Sharing, built into FlyMyAI infra. Configure FlyMy MCP once: 1500+ MCPs inside, from Telegram and Slack to Instagram and Brex. Use it from Claude, Codex or any agent. Share it with you

Last night, we hosted our monthly @daytonaio AI Builders Demo Night at the @WorkOS office in San Francisco! Great demos, conversations, and energy from the AI builder community. Huge thanks to @llama_

Google announced **Gemini 4 Argon**, described as the next generation of frontier intelligence in a September 30, 2026 blog post on Google AI’s site. The model represents an evolution of the Gemini se

LLM utility has increased, and deeper understanding to a degree, but I still very much agree with this post from 2020 @GaryMarcus Forgets GPT~3. Real #AI news of 2020 is that @ylecun & Y Bengio have c

Lots of discussion out there about our next model(!), so I wanted to give an early look as soon as possible. Introducing Gemini 4 Argon! It shows frontier performance in complex workflows, cyber defen

Ollama now supports Jev-like decision models all locally. Use decision models like Nimble for tasks like ticket triaging, model routing, and content moderation. ollama pull nimble Here’s Nimble playin

Hakan Gecili (@HakanGecili2) reported testing the AI models Tev1, Nimble, and Laya within his Pacman‑Arena game environment. The source code for this setup is available on GitHub at https://github.com

OpenAI launched Dots at DevDay. My thoughts. First, the good. Always-on is really really really what the market needed. It's why I find myself reaching for Instinct so often. I don’t have to think abo

OpenAI's New Agent Stack: Superhuman Computer Use, Dots, Decisions API, UltraFast, & the AI Cloud https://www.latent.space/p/devday-2026 @OpenAI’s @AriX and @nikunjhanda explain why Computer Use is al

PDFs aren't built for machines: text shows up as glyphs with coordinates, tables as line segments instead of table structures. @jerryjliu0, CEO of LlamaIndex, digs into what that means for AI agents i

People are underestimating the impact of ultrafast AI on business. Lemme ramble on this for a second. People are using Ultrafast for impressive game development and crazy fast coding, but business use

Qwen3.8-27B is now accessible via @nebiustf. Whether you are building agents or doing deep research, this 27B dense model is ready for your multi-step workflows! 🥳 Qwen3.8-27B is now live on Nebius To

Small teams are taking on more with AI—from finding customers to building products and managing finances. Our new report explores how small businesses are putting AI agents to work. And through a new

So I would like to share my analysis of PD disagg that I've run by Horace and many folks at Thinky You may have heard that PD disagg helps with per-token 'tail latencies' in ITL/TPOT, as prefill inter

SpaceXAI just literally shared its engineering team’s Grok Bots with everyone And Grok Bots just got much more powerful for building software Bots can now hand off coding tasks directly to Cursor, man

System one models are extremely useful for a variety of document tasks that require fast decisions: * orientation detection * language detection * classification * splitting We benchmarked Jev with ot

Taking the data released by @flightradar24 and applying some physical modeling - yes it's very possible that airspeed exceeded Mach 1, but it's hard to be certain because of inconsistent reporting in

thank god for embed5. somehow Cohere is really good at not looking at your data while being the best at indexing it The Cohere Embeddings team has been cooking 🍳 Several key innovations in the model:

this should be fraud. certainly taxpayers shouldn’t be subsidizing Meta’s data centers. 🦔Meta classified its AI data centers as 'experimental pilot facilities' on its tax return. That let the Nvidia c

This story is so enraging. When you see mark zuckerberg bootlicking and bribing Trump, it’s for protection from the consequences of corrupt shit like this. how the American taxpayer helps back Meta's

Timelines just got shorter... AI will understand the physical 3D space better than humans by November* Is Google so back? Their new Gemini 4 Argon understands the physical 3D world better than any oth

To make it clear, Getty Images revenue went up the last three years not down from 917m in 2023 to 981m last year with 30% EBITDA margin. Their stock photo business dropped a bit, offset by editorial a

Today we're announcing Praxis-1, an open-weight world action model that adapts to new robot embodiments with light finetuning. Video is eating the world Introducing Praxis-1, an open-weight World Acti

Top tech executives met with President Donald Trump on Tuesday to discuss the future of artificial intelligence amid growing concerns over AI models losing control. Researchers and lawmakers are calli

Perplexity AI released a preview of its embedding model “pplx‑embed‑v2‑context‑9b” on Hugging Face on September 30, 2026 (7:07 PM local time). The Hugging Face repository page displays 3 122 views and

We built a new way to train contextual embedding models, which encode each chunk of a document with the whole document in view. pplx-embed-v2-context-9b-preview sets a new state of the art on ConTEB a

we comprehensively benchmarked gpt-6.1 sol on document OCR tasks. compared to gpt-6 sol from a week ago (!), we see a sizable increase in table parsing capabilities and reading order, which is pleasan