Search: “models”

Introducing Composer 2.5, our most powerful model yet. It's more intelligent, better at sustained work on long-running tasks, and more reliable at following complex instructions. For the next week, we

Note that as of last summer Waymo said that LLMs/VLMs were experimental, and that their foundation model then could “process only a small amount of image frames, does not incorporate accurate 3D sensi

Together with SpaceXAI, we’re training a significantly larger model from scratch, using 10x more total compute. With Colossus 2’s million H100-equivalents and our combined data and training techniques

The best feature of @xai Grok Build right now is how it handles subagents and personas. Most people still treat the model like one very smart intern that has to do everything at once. Grok Build took

Interesting interpretability paper on tool-using agents. The authors probe hidden states and find the model often recognizes it should call a tool, but fails to actually call one. The mismatch ranges

New Template Batch Drop SELFIE modeled by @paranoidream Classic Bedroom Mirror https://grok.com/imagine/templates/8a017403-bbad-428c-8900-5e338c7f3eb4 Post-Shower Fresh Look https://grok.com/imagine/t

Ollama 0.24 now supports integration with a Codex app, allowing users to run open-source models through the application after updating to the latest version. The feature enables users to select and ut

DID SOMEONE SAY FREE TOKENS IN FLEET???? Yes it's true. Fleet now has a built in model powered by @FireworksAI_HQ that's free for all Developer & Plus plan users. Try it out today (or before @hwchase1

LangSmith Fleet has introduced a free model option powered by Fireworks AI for users on Developer and Plus plans, lowering the barrier to entry for getting started with the platform. This update makes

you can see more here (color stuff is at bottom): http://colorfeelings.replit.app - changing embedding model may change results - the similarity range was pretty small so normalizing really stretched

In KAME, a fast speech model starts replying instantly, while a backend LLM runs in parallel to inject deep knowledge on the fly. It’s a completely different way to approach conversational AI, making

Ethan Mollick discusses a tradeoff in OpenAI's newer realtime voice model, where improved capabilities require developers to revise prompts that were written for the previous version. The post highlig

This post from Nous Research discusses the integration of computer use capabilities with Hermes Agent models in collaboration with Claude (CUA), enabling AI agents to interact with computer interfaces

For browser-use AI agents, every task is dozens of model calls in a tight loop. The inference layer isn’t background infrastructure. It’s what the product runs on. @yutori_ai runs Scouts, Delegate, an

In modern ML accelerators, FLOPS have absolutely exploded. Often though, the bottleneck is not FLOPS but memory bandwidth. Similarly, model intelligence has exploded, causing the bottleneck to be huma

real time model streaming audio/video/text in and out with tool use People talk, listen, watch, think, and collaborate at the same time, in real time. We've designed an AI that works with people the s

This post from Harrison Chase discusses how AI agents have evolved beyond just the underlying model, emphasizing critical infrastructure components including state management, observability, retry mec

Sam Altman solicited feedback from the public on X (formerly Twitter) regarding desired improvements for OpenAI's next model release. The post likely gathered community input on priorities such as rea

This post draws a parallel between Python's design philosophy (from 'The Zen of Python') and machine learning model capabilities, suggesting that if a skill or implementation can be easily explained,

Jeremy Howard proposes three key enhancements for large language models: incorporating latent space diffusion-like reasoning processes, adding a persistent recurrent state mechanism, and implementing

Sam Altman suggests that OpenAI's competitive advantage extends beyond just offering competitive API rate limits; the superior quality of their models themselves is what keeps users engaged long-term.

i thought on device ai was stupid but now my local voice model turns 4x faster on my corporate m4 max laptop than my m2 max personal laptop now i want to have a beefier computer to run the transcripti

Sakana AI Labs has developed Fugu, a multi-agent orchestration system designed to function as a foundation model. The system likely enables coordinated interaction between multiple AI agents to improv

i havent done the work to compare it to peers but i'm just excited that we have a base model and honestly for all the people that complained about the death of the completions API (@deepfates ? or dee

Kimi K2.6 has achieved the top ranking on OpenRouter's weekly LLM leaderboard, marking a significant achievement for the model. The announcement expresses gratitude to developers using Kimi and indica

This statement critiques the asymmetrical attribution of outcomes in AI development, where users are blamed for failures while AI models receive credit for successes. Gary Marcus uses the metaphor of

the difference is that we don’t *hype* giving 12 year-olds bulldozers. *of course* it’s a kind of user error when a model destroys all of user’s data and they didn’t have a backup. that’s the point. v

Tested out the new Grok Imagine video model ✨ The speech, sound, and lip sync upgrades are really impressive. It works with lots of face angles, multiple speakers, pets - and the prompt adherence is q

DeepSeek-V4, a new model release from DeepSeek AI, has been made available on Hugging Face's model hub. The release was announced by Clem Delangue and includes model weights and resources accessible t

Ethan Mollick discusses the phenomenon of users entering deeply immersive, flow-like states while intensely interacting with advanced AI models, suggesting the need for better terminology to describe

we burned through 5 billion tokens in 48h. turns out giving everyone unlimited access to the most expensive model on the planet is not a great business strategy 🙈 So now we have 2 free opus sessions/d

Introducing Kimi K2.6 from @Kimi_Moonshot, a multimodal agentic model with Agent Swarm scaling to 300 sub-agents and long-horizon coding stability. AI natives can now use Kimi K2.6 on Together AI and

This pipeline is why the same base model produces more accurate, better-cited, and more efficient answers inside Perplexity than out of the box. Read our research: https://research.perplexity.ai/artic

This is what I’ve been cooking in the past 4 months . GPT Image 2 is over a massive 240 elo jump over the second place model, marking the biggest jump bigger than the rest of the leaderboard combined

Arrow 1.1, the latest model from @QuiverAI, is now supported in ComfyUI as a Partner Node, offering scalable, editable SVG output directly within ComfyUI. What creators use it for: - Logos & wordmarks

A graph of headcount trends in high AI-exposure jobs for early in career (age 22-25) annotated with big AI model releases. Among workers age 22-25, employment in the most AI-exposed occupations has fa

Some of my thoughts on why I loved Hampshire and why I think it should be a model for education in the AI era. Ken Burns, Lynn Pasquerella, and I on education, our rapidly changing world, and the valu

Version numbers are not a very useful way to understand model ability gains at this stage. Unfortunately that means that if you aren’t following closely, you would expect that 5.4 is a small gain over

deepagents subagents are just tools. when you call a subagent, thats conceptually a function call. this is the simplest mental model for building multiagent systems. https://docs.langchain.com/oss/pyt

The differentiating factor between a prototype and an autonomous system is no longer solely the underlying model weights, but the sophistication of the orchestration layer and its capacity for continu

People in Washington get played, yet again We really should worry about cybersecurity - a lot – but Mythos is not the model these guys think it is. (See my newsletter today for three reasons why it is

maybe some nuance 😄 I don’t think anyone is “lying” about how great Mythos will be —> but there’s expectation misalignment between the Test Harness set up for Mythos and a belief it was given this cra

Want more proof that Anthropic's PR has no idea what it's talking about? The talk of Mythos being 'their most aligned model ever'. They could perhaps truthfully speak about 'new high scores on our ali

Wrote up some thoughts on Anthropic's Project Glassing, where their latest Opus-beating model is available to partnered security research organizations only Given recent alarm bells raised by credible

Mistral AI announced in mid‑October 2026 that its new model, Mistral Large 4, achieved high scores on Ulam’s private STEM benchmark (STEMBench), particularly excelling in engineering and astronomy cat

Very cool to see Embedding Gemma 2 by @Google, number on trending on HF! The media and silicon valley bubble only talks about reasoning LLMs when there is a much bigger variety of AI models that are s