AI volume day, no model releases, one Bitcoin patch

Six of sixteen releases come from AI tooling, but every one is plumbing: inference runtimes, coding agents, and SDKs getting point bumps, not new weights. Self-hosting logs four releases including a Syncthing release candidate and a Gitea point release. Nostr sees three minor updates. The one Bitcoin item, a Fulcrum maintenance release, and the lone Privacy entry, an Obtainium patch, round out a day with no security advisories and nothing urgent. Skim for anything running these stacks; nothing here demands immediate action.

Bitcoin#

Fulcrum 2.1.4. Maintenance release in the 2.x series; routine upgrade for Electrum server operators.

Nostr#

Granary 12.0. Breaking change strips author display name and linked images from microformats2 output; check downstream consumers before upgrading.
Divine 1.0.25. Follow the action.
Nostream Website 1.3.1. 899 3ed70e4 Thanks @Ferryx349!

Privacy#

Obtainium 1.7.1. Fixes self-hosted Gitea/Forgejo URL handling and cleans up install-state bugs; relevant for anyone tracking apps from custom sources.

Self-hosting#

OpenWrt 25.12.6. Sixth service release of the 25.12 stable line; routine maintenance update for routers on that branch.
ntfy 2.32.0. Adds passive cluster-node support, writes now route only through the leader; matters for ntfy.sh operators running multi-node setups.
Syncthing 2.1.7-rc.3. Release candidate adds GUI grouping for devices and folders; test before pushing to production syncs.
Gitea 28.2.0. Adds a builtin OAuth2 app for gitea-vscode and loosens mention-email restrictions for custom watches; mostly minor fixes otherwise.

AI#

Qwen Code 0.25.3. Adds optional model-generated layouts to Test Failure Explorer, with validation and fallback to the fixed view.
ComfyUI 0.42.1. Fixes crashes in the MOSS image template workflow; apply if that workflow has been unstable.
vLLM 0.32.0. Large release, 814 commits; worth checking changelog for breaking changes before upgrading production inference servers.
openclaw 2026.10.5. Improves self-healing updates across devices and platforms, fixing stranded installs from interrupted update processes.
llama.cpp b11724. Fixes CUDA flash-attention race conditions; relevant for anyone running GPU inference who's seen instability.
Claude Code Action 1.0.253

Read on Freedom.Tech: https://freedom.tech/posts/2026-10-19-bitcoin-daily-brief/ Get the weekly recap in your inbox: https://freedom.tech/subscribe/

Freedom.Tech is powered by Foundation. Catch the weekly discussion on Freedom Tech Friday via Ungovernable Misfits.

AI volume day, no model releases, one Bitcoin patch

Bitcoin#

Nostr#

Privacy#

Self-hosting#

AI#

Comments0