Meta acknowledged that its Muse AI app was heavily inspired by OpenClaw while also saying Muse was built from scratch. The company’s product head, Nat Friedman, credited using OpenClaw in January and said Meta aimed to scale Muse to billions of people. This confirmation redirects the debate from “coincidence” to deliberate product inspiration, while Meta adds no further details beyond the statement.
Paper2Agent converts academic papers plus their code and supplemental materials into interactive AI agents that can extract workflows and run the methods from the paper on new data. The team says it generated 22 validated AlphaGenome tools in about 45 minutes for under US$15 of compute cost. As a result, researchers can ask in plain English and get runnable analyses and visualizations, and multiple paper-derived agents can collaborate to propose and test hypotheses.
UNA Watch commercialised its modular, repairable sports GPS watch, shifting from crowdfunding to full shipping across 32 countries and aiming to challenge Apple and Garmin’s sealed-device model. The watch is priced at £179 in the UK. UNA is enabling battery and display swaps and publishing open hardware and an open software platform with an SDK so owners and third parties can build apps, aided by AI-assisted coding tools described in the article.
Kev, a GitHub repo for small Jev-like decision models, was released with local training and a TypeSafe System One–compatible API plus a web playground. The repo ships Kev models sized 0.8B, 4B, and 9B, including a note that Kev-4B and Kev-9B fit a 32 GB Mac. Users can run a local server, call /v1/systemone with multiple question types that return calibrated probabilities and confidence, and compare answer effects via permuting and isolation demos.
Gortex (a GitHub repo) shipped a local, no-dependency system that parses 257 languages into an in-process provenance graph and exposes it to coding agents via MCP, a CLI, and an HTTP/Web UI. It reports $168.69 in cost avoided (claude-opus-4) across 1,878 calls, saving 11,246,094 tokens against naive file reads. Developers can now use graph-based symbol lookups, call-chain and blast-radius queries, and AI-assisted PR review through the connected agents instead of repeatedly re-reading source files.
Open-source parsing and bitmap libraries were rebuilt and re-benchmarked, showing large performance gains across six widely used codebases in 2026 after long flat periods. simdjson’s serialization speed rose 2.1x on citm_catalog.json from February 2026 onward. The result is faster URL/number/JSON/unicode parsing and bitmap operations for many downstream users, likely enabled by more accessible optimization work in that period.
Jun Kim, creator and maintainer of oMLX, has joined Hugging Face to support the MLX local-AI community. The article says oMLX will remain under the Apache 2.0 license. This funding and full maintenance shift is expected to stabilize oMLX and speed up long-term development, while Hugging Face works to help transformers model definitions transition into reference MLX implementations.
Every AI story that matters,
in your inbox by 8am.
TLDRocket reads all relevant sources, removes duplicate coverage, and summarises the
day in two minutes. Follow companies and topics for alerts, or get the
briefing in Slack. Free, no spam, unsubscribe anytime.
Reading TLDRocket needs no cookies, and the readership counts we rely on come from
our own cookieless analytics. Google Analytics is the exception: it sets cookies and
reports to Google, so it stays switched off until you allow it. You can change your
mind any time from “Cookie settings” in the footer.
Strictly necessary
Session security and form protection (tldrocket-session,
XSRF-TOKEN, 2 hours). The site cannot work without them,
so they need no consent.
Always on
Google Analytics 4 (_ga,
_ga_<id>, up to 2 years). Measures which
stories and sections readers use. Google acts as a third-party processor and may
store the data outside the EU. No advertising, no profiling, no data sold.
