Meta’s Astryx Brings a CLI and MCP Server to an Open-Source React Design System Agents Can Read
Meta released Astryx this week.
AINOVAT
Source
60 articles from this source
Meta released Astryx this week.

In this tutorial , we explore the Open-SWE-Traces dataset as a practical resource for studying and preparing agentic software-engineering trajectories for fine-tuning.

A new Cursor study reports that newer coding agents often retrieve known fixes instead of deriving them, inflating popular benchmark scores.

Perplexity launched Computer for Counsel.

OpenAI has begun a limited preview of GPT-5.6 , its next-generation model series.
Apple research team recently released the container project .

Tutorial on building a lightweight AI agent inspired by nanobot architecture in Google Colab.

DeepReinforce has released Ornith-1.0 , an open-source model family built for agentic coding.
Most end-to-end OCR models slow down as output grows.

Gradium today released two real-time speech translation models: stt-translate and s2s-translate .

In this tutorial , we build OpenHarness from scratch to better understand how a practical agent harness works.

In this tutorial , we build a fully offline Graphify workflow that turns a realistic multi-module Python application into a knowledge graph.

Nous Research has expanded the Skills System inside Hermes Agent, its open-source self-improving agent.

Generative AI has reshaped how software gets built.

Autoregressive large language models generate text one token at a time.

Today, Mistral AI released OCR 4 , its latest document-understanding model.
Datalab has released lift , a 9B open-weights vision model for structured extraction.

In this tutorial , we build a speech recognition and translation workflow using NVIDIA Canary-1B-v2 .

Prime Intellect has released prime-rl version 0.6.0 .

In this tutorial , we work with GLM-5.2 and use its hosted, OpenAI-compatible API instead of running the full model locally.

xAI shipped a new mode called /goal inside Grok Build , its terminal coding agent.

Today, Sakana AI launched Sakana Fugu .

MoonMath AI team has released a bf16 forward attention kernel for AMD’s MI300X GPU.

In this tutorial , we build a Prefab application that demonstrates how to create interactive dashboards entirely in Python.

Large language models are stateless by default.

In this tutorial, we build a full Crawlee-for-Python workflow that covers environment setup, local website generation, static crawling, dynamic crawling, structured extraction, and downstream data processing.

Getting prompts right is still the hardest part of shipping reliable LLM applications.

Nous Research has added a Blank Slate setup mode to its open-source Hermes Agent.

Yandex has open-sourced YaFF ( Yet another Flat Format ) under Apache 2.0.

Tutorial on building an end-to-end forecasting workflow with TimeCopilot using foundation models and automated anomaly detection.

NVIDIA Research has released SpatialClaw, a training-free framework for spatial reasoning.

While recent breakthroughs in AI reasoning have largely been driven by massive scale, pouring in billions of parameters to cross complex cognitive thresholds— VibeThinker-3B is charting a completely different path.
This week, Liquid AI released two new retrieval models.

In this tutorial, we implement an end-to-end workflow for Salesforce CodeGen .

Most AI memory remembers the user.

KV cache compression methods, including TurboQuant, OSCAR, and EpiCache, aim to reduce memory usage in long-context large language models.

OpenAI's LifeSciBench evaluates AI models on real life-science research tasks with expert-written rubrics.

Vercel has released eve , an open-source framework for building, running, and scaling agents.

MiniMax releases MSA, a sparse attention method built on Grouped Query Attention, targeting the quadratic cost of softmax attention at long context.

OpenAI publishes Deployment Simulation, a new pre-deployment safety method that simulates model deployment to assess risks before release.

In this tutorial , we implement xFormers : a practical toolkit for building fast, memory-efficient Transformer models on GPUs.

The Qwen team has released three embodied AI models, grouped as Qwen-Robot-Suite.

Nous Research has shipped a change to Hermes Agent.

The concept of vibe coding is interesting; you don’t need to be a developer or software engineer to build your own applications.

Foundation models keep getting stronger, yet they still stall on the same thing: context.

In this tutorial, we build a workflow for using Docling Parse to analyze PDF documents at a detailed structural level.

Tokyo-based Sakana AI shipped its first commercial product ‘Sakana Marlin’ this week.

k-means has been an offline tool for decades.

GLM-5.2 is the latest large language model from Z.ai, becoming the third major release in the GLM-5 line.

In this tu t orial , we explore the FineWeb dataset through an advanced hands-on workflow.

Databricks open-sources Omnigent, a meta-harness for composing, governing, and sharing AI agents across multiple platforms.

In this tutorial, we implement a QwenPaw workflow that provides a practical environment for building and testing an agent-powered assistant.

Anthropic has disabled its two most capable models for every customer.

This week, Moonshot AI released Kimi K2.7-Code .

Google Research team has announced the launch of Gemini-SQL2 on X .

Moonshot AI has introduced Kimi Work, an AI agent that runs on your own desktop.

Zyphra has released Zamba2-VL, a family of open vision-language models.

In this tutorial, we build an end-to-end 3D medical image segmentation pipeline using MONAI to segment the spleen on the Medical Segmentation Decathlon Task09 dataset.

Perplexity has moved Deep Research into Computer, its multi-model orchestration system.

Today, xAI shipped the Grok Build Plugin Marketplace.