Runway releases its robotics model Praxis-1 with open weights and source code under the Apache 2.0 license. The model uses knowledge from video training for robot control and leads the World Action Model category on Nvidia's RoboLab-120 benchmark. It can't yet plan complex action sequences on its own, though.
The Universal Agent Evaluation Framework (UAEF) is a framework-agnostic system for evaluating AI agents. UAEF converts each framework's traces into one canonical format, scores them with 36 built-in metrics across 7 dimensions, and stores results so teams can compare agent versions and catch regressions.
Research: parse_number() as a SQLite extension: C vs Rust I was out walking the dog when I got curious about how easy/difficult it would be to write a SQLite C extension that adds a parse_number() function for turning strings like -$1,234.56 or '12.5%' into numeric values that could be summed and sorted.
Ricursive Intelligence is releasing Ripple-65B Preview, a 65-billion-parameter open-weight language model that the company says achieves the top score among open-source U.S. models on the TerminalBench 2.1 coding benchmark at a total R&D cost of $32 million. On other benchmarks, however, Ripple falls behind the likewise newly released western open model Inkling 2 Small.
Today we're introducing a framework for creating these environments; we call them FoundationWorlds. Our first foundation world is ToyBlocks : an open simulation framework built around a robot arm and blocks on a tabletop.
NVIDIA Cosmos world model development platform adds four new open-weight checkpoints to its Cosmos 3 model family for generating interactive world simulation and multi-sensor multi-view driving scenes. All four models are post-trained examples of Cosmos 3 Nano and have 16B trainable parameters. The models use OpenMDW1.1, permitting commercial and non-commercial use under its terms.
Release: llm 0.37.1 A bugfix release for LLM , with contributions from Steven Huang , Tao An , Arjan Mossel , and XQ .
Release: datasette-llm 0.1a10 Fixes a bug that affected my agent.datasette.io demo site, where the set of available LLM models was not correctly restricted to just those in the configuration.
On July 30, 2026, Transformers enabled PyTorch's scaled dot-product attention (SDPA) for T5. As a side effect, UMT5's decoder could attend to future tokens.
Release: llm 0.37 There's a lot of stuff in this new release of my LLM CLI tool and Python library, representing around two months of accumulated features and fixes. I added gpt-6.1-sol and gpt-6.1-terra , plus a new llm openai endpoint add command to make it easier to configure additional models that use OpenAI-compatible endpoints.
Y Combinator Garry Tan has jokingly admitted that he's suffered from "cyber psychosis” and barely sleeps sometimes because he’s so excited to be working with his AI agents. On a new episode of Walter Isaacson’s podcast, The Builders, Tan put a price tag on his passion.
dbreunig/nesformer Inspired by qwenputer by Divyansh Agrawal, Drew Breunig baked a working emulator for a Nintendo Entertainment System 6502 CPU into the weights of Qwen3.5-4B. You can now play NES ROMs such as Super Mario Bros (very slowly) running inside Qwen!
Mistral says its new model can afford organizations a kind of control over its cyber defenses that closed models can't.
Nvidia has released Nemotron 3.5 Transcribe, an open speech recognition model that handles transcription, speaker recognition, and translation and takes its instructions as text prompts. It leads several benchmarks and uses only a sixth of the active parameters of its strongest open rival.
Nvidia's new data engine Flora labels one hour of robot video for 72 cents instead of $88 with GPT-6 Astra, while scoring slightly higher on quality. The open system, built on three fine-tuned Qwen3.5 models, catches errors in robot actions far more reliably than any competitor but sometimes falls short on descriptive accuracy.
Four open Granite models, one extensible forecasting interface, and less guesswork about which model to use. One recipe combines four complementary Granite architectures and returns median and quantile forecasts, with no retraining and no fitted ensemble weights.
SimReady Central Library brings together more than 2,000 simulation-ready (SimReady) OpenUSD assets on Hugging Face, giving developers a free, ungated starting point for physical AI training environments and AI factory digital twins.
Last November, we introduced AnyLanguageModel on this blog: a drop-in replacement for Apple's Foundation Models framework that lets you choose the model and provider your app uses. A year and dozens of releases later, AnyLanguageModel 1.0 is out, along with swift-huggingface 1.0 and swift-xet 1.0.
Release: llm-echo 0.5 I'm working towards a new LLM release, and testing PRs caused me to revisit the llm-echo debug plugin. This plugin provides a "model" which doesn't involve an LLM at all - it just repeats back the prompts that have been fed to it.
Release: llm-tools-datasette 0.1.1 When I switched LLM over to httpx2 in LLM 0.33 I forget that this would break any plugins which used httpx without directly depending on it. llm-tools-datasette was one of those plugins, now fixed.
Beam by Reflection is out on Hugging Face! It is the first model from the new US lab, Reflection, and it’s open! It's a mixture-of-experts (MoEs) model with 501B parameters (23B active per token), supports text as input, has a context window of 1M tokens, and is released under the Apache 2.0 license.
Research: sparqlite: SPARQL 1.1 inside SQLite, written in Rust This research task is really an experiment in running Claude Code for web in "Ultracode" mode, as described in Orchestrate subagents at scale with dynamic workflows in the Claude Code documentation.
Black Forest Labs FLUX.3 Image Klein is here, and it runs in Diffusers from day one. One model does text-to-image and multi-reference image-to-image, and it adds something new, the layout control.
How Leash Bio tamed DELs, the assay that traumatizes everyone, and used the data to catch cheating models. And why that same data is perfect for AI agents learning to think like chemists.
Release: web-edit 0.1a1 Includes performance optimizations written by Claude Opus 5.5 after I fed it this Python 3.15 profiling report generated by GPT-6 Astra.
Release: ttok 1.0 I released ttok 0.4 , ran uv tool upgrade ttok , piped a file into the new version... and realized that it was defaulting to the GPT-4 tokenizer when it should very clearly default to GPT-5/GPT-6 instead! I figured switching the default was a reasonable excuse to finally ship a 1.0.
Release: ttok 0.4 ttok is my CLI tool for counting tokens, using OpenAI's open source tiktoken library. It hasn't been in updated in a couple of years, but I finally fixed a Click warning, updated CI, and added a --list-models command to list available models.
Release: datasette-atom 0.11a0 A minor fix for compatibility with the latest Datasette alphas. This meant we could upgrade the datasette.io site to Datasette 1.0a41.
Aleph Alpha has released Kolibri, a German-English mixture-of-experts model with 78 billion parameters, about three billion of which are active per token. Over 21 percent of the training data is German. The model was trained on 768 B200 GPUs in Germany and Finland, and the weights are freely available under the Apache 2.0 license.
Getting some extra storage space on your Mac could be as easy as deleting Apple's AI features with a new open-source command line tool called RemoveMacAI. There used to be a single Settings toggle for disabling Apple Intelligence, but that was removed in macOS 27.
NASA and IBM have released the Lunar Foundation Model, one of the first open-source AI models for lunar science. Trained on nearly 2 million tile bundles, mostly from 17 years of Lunar Reconnaissance Orbiter data, it cuts the error in predicting polar ice deposits by up to 22 percent compared to the strongest model it was tested against.
We built AstaBrief 8B, a model that turns a research question and retrieved literature excerpts into a cited report. AstaBrief is available in Asta's Generate a report feature today as Fast mode alongside Claude-powered Thinking mode, and we're also open-sourcing it and the training data so other
Release: commit-rewriter 0.2 Support for branches other than the default branch. Use uvx commit-rewriter --branch other to run against another branch. #3
Release: llm-typesafe 0.1a0 I built this new plugin for LLM to add support for TypeSafe AI's new Jev model .
Custom-made molecules are advancing medicine, materials, and agriculture, but producing them is slow and expensive. A new Nature paper highlights RetroChimera, a predictive model that helps accelerate chemical synthesis, helping researchers explore a wide range of molecules.
Hugging Face released tokenizers v1, preserving token IDs, APIs, vocabularies and merge ranks from v0.23 while increasing speed by up to tens of times.
Google Research released MilleMiglia, a C++ generator of realistic, privacy-preserving benchmark instances for middle-mile logistics delivery. Source and documentation are public on GitHub.
Children's Hospital of Philadelphia (CHOP) uses MONAI, the open-source medical imaging framework co-founded by NVIDIA, to reduce paediatric heart modelling from four hours to seconds, already supporting complex ventricular septal defect surgical planning.
Hugging Face launched NeoMME in 260M and 800M sizes. A single bidirectional Transformer handles text tokens and 32×32 image patches, trained from scratch with a masked discrete diffusion objective rather than pretrained vision towers or causal language models.
A public low-cost approach fine-tunes LFM2.5-350M with TRL GRPO using around 500 samples and 100 steps, raising IFStruct from 22.6% to 29.7%. Training fits free Colab or Kaggle GPUs, with local llama.cpp evaluation on a MacBook and code open on GitHub.