model releases

arXiv:2605.13989v1 Announce Type: new Abstract: We present VectraYX-Nano, a 41.95M-parameter decoder-only language model trained from scratch in Spanish for cybersecurity, with a Latin-American focus

arXiv:2012.14425v2 Announce Type: replace-cross Abstract: Cyberattacks cause billions of dollars in damage annually, with malicious hackers often sharing exploit code and techniques on underground for

arXiv:2510.05213v2 Announce Type: replace-cross Abstract: Pretrained vision foundation models (VFMs) advance robotic learning via rich visual representations, yet individual VFMs typically excel only

arXiv:2605.14542v1 Announce Type: new Abstract: A skilled live-commerce host is not merely a narrator, but a sales agent who converts viewer curiosity into purchase intent through expert product knowl

arXiv:2605.14607v1 Announce Type: new Abstract: Any new medium, once it emerges, is used for more than the transmission of overt content alone. The information it carries typically operates on two lev

arXiv:2605.13923v1 Announce Type: cross Abstract: We study certified runtime monitoring of past-time signal temporal logic (ptSTL) from visual observations under partial observability. The monitor mus

arXiv:2602.07045v2 Announce Type: replace-cross Abstract: Recent advancements in Multimodal Large Language Models (MLLMs) have enabled complex reasoning. However, existing remote sensing (RS) benchmar

We just added significantly more NVIDIA Blackwell GPUs to better serve GLM-5.1 model on Ollama's cloud. We have been adding more GPUs daily for all the other models. Claude Code: ollama launch claude

arXiv:2605.14290v1 Announce Type: cross Abstract: ReAct has become the default architecture across LLM agents, and many existing web agents follow this paradigm. We argue that it is the wrong default

Weekends are for vibe coding. But are your vibes continuously improving? Fine-tune your own model → stop waiting on someone else's release cycle. Today's training update: Gemma 4 Dense is now availabl

Written by: Austin Larsen, Tyler McLellan, Genevieve Stark, Dan Ebreo Introduction Google Threat Intelligence Group (GTIG) has continued to track an expansive extortion campaign by UNC6671, a threat a

We've created comprehensive documentation of the latest X algorithm via DeepWiki: https://deepwiki.com/xai-org/x-algorithm Engagement parameter weights are still private (so we don't know the relative

We've gone even farther: Nemotron 3 Super is 120B and pretrained on 25T tokens in NVFP4. Nemotron 3 Ultra is ~500B and also pretrained in NVFP4. Accelerated computing means we rethink every aspect of

arXiv:2603.07880v5 Announce Type: replace Abstract: Moltbook is the first large-scale social network built for autonomous AI agent-to-agent interaction. Early studies on Moltbook have interpreted its

arXiv:2605.14478v1 Announce Type: cross Abstract: Context: Retrieval-augmented code generation relies on cross-file repository context, but retrieved snippets may come from obsolete project states. Ob

arXiv:2605.14504v1 Announce Type: new Abstract: Long-horizon household tasks demand robust high-level planning and sustained reasoning capabilities, which are largely overlooked by existing embodied A

This entry documents a Wordle game result (puzzle #1,790) played by Anthropic, showing a failed attempt marked as X/6, meaning the player did not solve the puzzle within six tries. The color-coded emo

This post documents a Wordle game result where the player successfully solved puzzle #1,791 in four attempts, using color-coded emoji feedback (black for incorrect letters, yellow for correct letters

Carmen Arroyo / Bloomberg: xAI launches Grok Build, an agentic CLI for coding, building apps, and automating workflows, in beta for SuperGrok Heavy subscribers — Elon Musk's xAI is rolling out its fir

arXiv:2605.14754v1 Announce Type: new Abstract: Large Language Models (LLMs) are increasingly deployed for knowledge synthesis, yet their capacity for compositional generalization in scientific knowle

This post discusses how traditional caching mechanisms assume deterministic behavior (identical inputs producing identical outputs), an assumption that breaks down with large language models and espec

arXiv:2605.12586v1 Announce Type: cross Abstract: Vision-language models (VLMs) exhibit a striking paradox: they can generate executable code that reconstructs a 3D scene from geometric primitives wit

4/5 The fix: stop encoding execution order, encode graph position instead. Now you get two independent controls: experiment name (controls the key) and component code (controls the prompt). Change eit

5/5 What this unlocks: - Change one prompt, serve cached results for all other components. - A/B test any component against identical upstream outputs. - Run best-of-N inference without N branches col

arXiv:2603.14066v2 Announce Type: replace-cross Abstract: Many real-world multi-party negotiations unfold as sequences of binding, action-level commitments rather than a single final outcome, yet this

arXiv:2605.13015v1 Announce Type: cross Abstract: The geometry of the retinal vessel is a key biomarker of vascular diseases, yet clinical evidence remains primarily observational. Existing generative

A small hack for my AI multi-taskers out there: I have so many Claude Code conversations running at once that I ask it to synthesize all of its action or my next action into single go/no-go sentences.

arXiv:2605.12685v1 Announce Type: cross Abstract: Graph Self-Supervised Learning (GSSL) has emerged as a powerful paradigm for generating high-quality representations for graph-structured data. While

arXiv:2505.12942v4 Announce Type: replace-cross Abstract: Large language models have demonstrated remarkable performance; however, their massive parameter counts make deployment highly expensive. Low-

arXiv:2605.12622v1 Announce Type: cross Abstract: We formalize action emergence as a target capability for end-to-end autonomous driving: the ability to generate physically feasible, semantically appr

arXiv:2605.12816v1 Announce Type: new Abstract: The Average Gradient Outer Product (AGOP) governs feature learning in neural networks: the Neural Feature Ansatz states that weight Gram matrices at eac

arXiv:2605.13532v1 Announce Type: new Abstract: As generative AI (GenAI) tools become easily accessible, there is promise in using such tools to support instructors. To that end, this paper examines u

This AINews roundup from Latent Space likely covers updates on OpenAI's Codex model gaining prominence in code generation applications, alongside news about Anthropic's Claude implementing usage meter

arXiv:2601.16806v3 Announce Type: replace Abstract: In this work we develop a novel insect-inspired model for visual point-goal navigation. This combines abstracted models of two insect brain structur

Nous Research announced the release of a new model available for free to Hermes agent users through the Nous Portal. The post suggests this is a significant model release from Nous Research, their org

Anthropic PBC, the developer and provider of the Claude artificial intelligence model family, said it’s offering a special credit pool for users who want to use agentic tools with its large language m

Anthropic PBC today launched an offering called Claude for Small Business that is designed to expand the adoption of its chatbot among entrepreneurs. The feature bundle is the fifth market-specific ve

any time a model router company drops data, its worth browsing. here we learn that gemini leads in education and personal assistants (?!), ant leads in vibecoding and koding and back office (?!), and

arXiv:2605.12879v1 Announce Type: new Abstract: Doubly-stochastic attention has emerged as a transport-based alternative to row-softmax attention, with recent Transformer variants using it to reduce a

arXiv:2605.13548v1 Announce Type: cross Abstract: Existing robotic foundation models, while powerful, are predicated on an implicit assumption of temporal homogeneity: treating all actions as equally

arXiv:2605.13784v1 Announce Type: new Abstract: Conventional transformer inference engines are request-driven, paying an O(n) prefill cost on every query. In streaming workloads, where data arrives co

Automating AI research is the next major step in AI We let Claude Code (Opus 4.7) and Codex (GPT 5.5) run autonomously on the nanoGPT speedrun optimizer track using our idle compute. ~10k runs, ~14k H

arXiv:2605.13455v1 Announce Type: new Abstract: Synapses are densely packed submicron structures that dynamically reorganize during learning and memory formation. Longitudinal extit{in vivo} imaging o

arXiv:2605.12843v1 Announce Type: cross Abstract: Model merging aims to combine multiple task-specific expert models into a single model without joint retraining, offering a practical alternative to m

arXiv:2409.02038v3 Announce Type: replace-cross Abstract: Existing text-to-SQL benchmarks have largely been constructed from public databases with well-structured schemas and simplistic question-SQL p

arXiv:2512.18951v3 Announce Type: replace Abstract: Infants learn not only object categories but also fine-grained visual attributes such as color, size, and texture from limited experience. Prior inf