Search: “model-releases”

arXiv:2602.16571v2 Announce Type: replace Abstract: Large-scale sharing of dialogue data is key to advancing the science of teaching and learning, yet rigorous de-identification remains a major barrie

This is a release candidate version of Ollama that merges updates from the main development branch into the llama-runner-phase-0 branch, likely incorporating bug fixes and features in preparation for

arXiv:2605.07872v1 Announce Type: cross Abstract: Multimodal reward models have advanced substantially in text and image domains, yet progress in video understanding reward modeling remains severely l

arXiv:2605.06708v1 Announce Type: cross Abstract: Visual text compression (VTC) promises efficient long-context processing by rendering text into an image and re-encoding it with a vision-language mod

arXiv:2605.06175v2 Announce Type: replace Abstract: Vision-language-action (VLA) models inherit rich visual-semantic priors from pre-trained vision-language backbones, but adapting them to robotic con

📣We're calling for ambassadors! Whether you're a developer with great technical taste or a local community leader who loves bringing people together, we'd love to have you join us. Visit the website b

OpenAI announced the acquisition of Tomoro, a company that will add 150 experienced Forward Deployed Engineers and Deployment Specialists to OpenAI's team. This acquisition appears focused on expandin

arXiv:2605.07260v1 Announce Type: cross Abstract: Mixture-of-Experts (MoE) language models route each token to a small subset of experts, but whether the routes selected by a trained top-k router are

arXiv:2605.06772v1 Announce Type: new Abstract: As large language models (LLMs) show increasing promise on research-level physics reasoning tasks and agentic AI becomes more common, a practical questi

arXiv:2605.06731v1 Announce Type: cross Abstract: Personalized LLM agents maintain persistent cross-session state to support long-horizon collaboration. Yet, this persistence introduces a subtle but c

arXiv:2605.07114v1 Announce Type: new Abstract: Reinforcement learning with verifiable rewards (RLVR) has emerged as a central paradigm for improving the reasoning capabilities of large language model

arXiv:2605.07984v1 Announce Type: cross Abstract: We study planning site formation in language models -- where internal representations of structurally-constrained future tokens form during the forwar

arXiv:2605.07527v1 Announce Type: cross Abstract: Recent work has observed that explanations produced by Self-Interpretable Graph Neural Networks (SI-GNNs) can be self-inconsistent: when the model is

arXiv:2605.07068v1 Announce Type: cross Abstract: The LLM Wiki pattern, to compile and provide domain knowledge into a persistent artifact and serve it to LLMs via KV cache inference, promises context

This post from Elon Musk's X account likely recommends or highlights content that Musk finds valuable or important, though without access to the specific tweet content, the exact subject matter cannot

would you call it a superapp? After being a Claude Code devotee for a year, I finally tried Codex on a new project this weekend. Once again, in the matter of a few months, it feels like the world chan

arXiv:2605.07579v1 Announce Type: cross Abstract: Reinforcement learning with verifiable rewards (RLVR) for Large Reasoning Models hinges on baseline estimation for variance reduction, but existing ap

Anthropic says ‘evil’ portrayals of AI were responsible for Claude’s blackmail attempts https://techcrunch.com/2026/05/10/anthropic-says-evil-portrayals-of-ai-were-responsible-for-claudes-blackmail-at

Apple may be planning to role out its updated Siri based on 2024's vision at the moment when Claude Code and Codex (also OpenClaw) can increasingly do the actual assistant thing: read my emails & cale

Deep Agents Deploy! https://docs.langchain.com/oss/python/deepagents/deploy Claude Managed Agents is really good But we need an open source solution Haven’t found any options so might need to build my

This post provides a detailed review and practical guide for setting up and testing Ollama with DeepSeek models specifically on Ryzen APU systems. It likely covers performance benchmarks, configuratio

@eladgil BS. Attention was born in Montréal PyTorch in NYC. AlphaGo in London AlphaFold in London ESMFold in NYC Llama 1 in Paris. Llama 2 in Paris+NYC+SV DeepSeek in Hangzhou Plus: DINO in Paris JEPA

Jarred tried rewriting Bun in Rust and it passes 99.8% of the existing test suite we're not being ambitious enough there will be a blog post about this. on what this means for bun, benchmarks, memory

Local AI is having its moment! Below is the number of new GGUF models created each month over the past 8 months & insights from our HF internal agent (May is partial): - 176,000 total public GGUF mode

PLEASE DO NOT PANIC about the Mythos/METR graph that everyone is panicking about. Progress is being made but people are totally overreacting. Here’s some context that is being left out from nearly eve

Project Tapestry by @thealliance_ai: gathering some of the best minds in the world in Paris, to help solve the problem of AI Sovereignty for Viet Nam (and Japan and India and Thailand and France and S

Sam Altman comments on a statement by @roydanroy, suggesting that the technological capabilities or developments being discussed would not have been possible prior to GPT-5.5. The post implies that GP

The personification of Claude — in name (the only AI with a human one), in training, in Anthropic’s philosophy (see Claude Constitution), in fanfiction (see the Claude cartoons), etc — feels quite con

'The Spaghetti Benchmark' refers to the meme of Will Smith eating spaghetti, which became shorthand for the unhinged early stage of AI video-generation. The original 2023 clip generated with ModelScop

This is just mindblowing stuff! I couldn't resist replicating this workflow to generate 3D biological structures. In a few minutes, I designed an artifact specifically built to generate these for any

This post from Anthropic's X account shares a completed Wordle puzzle (puzzle #1,785) solved in 4 attempts, displaying the color-coded guess feedback that indicates which letters were correct, misplac

This appears to be a Wordle game result post from Anthropic's X (Twitter) account, showing completion of puzzle #1,786 in four attempts with a specific pattern of correct (green), misplaced (yellow),

100% agree on the Context Hub. Developers constantly tweak their approaches to manage context in their prompts with each new model and tool suite release. I would even say that the core problem we fac

Kate Clark / Wall Street Journal: A profile of Anthropic CFO Krishna Rao, who tends to take a conservative approach to revenue projections and has chosen to raise less money than is available — Krishn

Anthropic: Anthropic details how it improved Claude's safety training after finding agentic misalignment in older models, such as Opus 4 blackmailing engineers — Last year, we released a case study on

BeeLlama.cpp is an optimized implementation featuring advanced DFlash and TurboQuant quantization techniques with support for reasoning and vision capabilities. The project demonstrates running Qwen 3

DeepSeek released the full technical report for DeepSeek-V4 on April 24, 2026, titled 'DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence.' The paper details FP4 quantization-awa

Downloading now... 1M token context window with supposedly usable coding agent capability all on a 128GB Macbook Pro is 🤯 🚨 OPEN SOURCE AI IS LITERALLY UNSTOPPABLE 🚨 The legendary founder of Redis (An

ERNIE 5.1 is here 🚀 ERNIE 5.1 significantly reduces pretraining cost while compressing total parameters to ~1/3 and activated parameters to ~1/2 — using only ~6% of the pretraining cost compared to mo

Hot take on METR’s new graph that so many people are flipping about today. • Claude Code is a real advance; Mythos probably builds on some of what is learned there. But… • If you read the graph carefu

Huh. We evaluated an early version of Claude Mythos Preview for risk assessment during a limited window in March 2026. We estimated a 50%-time-horizon of at least 16hrs (95% CI 8.5hrs to 55hrs) on our

I recently installed Hermes after raw-dogging Claude Code for a year or so. Have tried it at work, didn't see the need to implement it personally. Always had the usual fears about 'privacy' and what n

If you want to look under the hood at the actual custom CUDA kernels and see exactly how we implemented the TwELL format for H100 GPUs, we’ve released the reference code. GitHub: https://github.com/Sa

My take on the Dawkins/Claude/Claudia/Claudius kerfuffle in @thenerve_news - with many 🙏🏽to @GaryMarcus for reminding me how Dawkins wrote so powerfully about the weakness of arguments from personal i

Kirsten Korosec / TechCrunch: NHTSA says the 2026 Tesla Model Y is the first car model to pass the agency's new ADAS tests; Tesla conducted the tests and submitted the results to the NHTSA — The Natio

An iOS application that enables large language models to run directly on-device using llama.cpp technology, allowing users to integrate their own Ollama instances for processing Apple HealthKit data t