• Bitcoin
• Lightning
• Nostr
• Freedom Tech
• Privacy
• AI

§Bitcoin

• btcd 0.26.0-beta.rc1. What’s Changed rpcclient: ensure http dial respects timeout by @wydengyre in https://github.com/btcsuite/btcd/pull/2450 fix spelling error by @jlopp in https://github.com/btcsuite/btcd/pull/2468 build: bump version of…

§Lightning

• Blink Mobile 2.4.45. Features - LNURL pay and Lightning Address support for self-custodial sends ( 3765) - Stable-sats receive auto-convert and send-flow fixes ( 3764) - Add Stable Balance and BTC - USD conversion flow ( 3761) - Self-cust…

LNURL pay now works for self-custodial sends; self-custodial onboarding flow ships.

• Phoenixd 0.8.0. What’s Changed Upgrade lightning-kmp to 1.12.0 by @pm47 in https://github.com/ACINQ/phoenixd/pull/232 Full Changelog : https://github.com/ACINQ/phoenixd/compare/v0.7.3...v0.8.0 Verifying signatures You will need gpg a…
• Lightning Pool 0.7.0-beta. This release of pool ships a handful of reliability fixes around the auctioneer subscription stream and pending-channel handling, plus the usual round of dependency bumps.

Reliability fixes for auctioneer subscription stream and pending-channel handling.

§Nostr

• Amethyst 1.11.0. What’s Changed feat(profile): restore Lightning Address + LNURL fields in Edit Profile by @vitorpamplona in https://github.com/vitorpamplona/amethyst/pull/2947 Raises the default maximum hashtag limit from 5 to 8 by @…

§Freedom Tech

• Tails 7.8. Changes and updates Update Tor Browser to 15.0.14 .

§Privacy

• JoinMarket-NG 0.30.0. Major improvements for wallet performance, specially for exiting old wallets with lots of transactions.

Operators with old wallets holding many transactions should see faster load times.

• Bitwarden Clients CLI 2026.4.2. - Security fix for local storage of API keys - Various under-the-hood improvements and minor bug fixes

Security fix for local API key storage; upgrade recommended.

• Bitwarden Clients Desktop 2026.4.0. - Targeted performance improvements on read/write - Refactor unlock service to use Bitwarden SDK - Updates to prevent losing unsaved changes when creating a Send - Updated default clipboard clearing time to 5 minutes…

Default clipboard clear now five minutes; targeted read-write performance improvements.

§AI

• llama.cpp b9263. mtmd, model : merge HunyuanOCR into HunyuanVL and fix OCR vision precision ( 23329) - HunyuanOCR shares the same HF arch and vision layout as HunyuanVL butwas split into a separate path that skipped the +0.1 bilinear…

HunyuanOCR now merged into HunyuanVL path with correct bilinear sampler; relevant if running either model.

• langchain-fireworks== 1.4.0. Changes since langchain-fireworks==1.3.1 release(fireworks): 1.4.0 ( 37582) feat(fireworks): migrate to fireworks-ai 1.x SDK ( 37581) chore(model-profiles): refresh model profile data ( 37574) chore: bump idna from 3….

Migrates to fireworks-ai 1.x SDK; breaking change if using older SDK version.

• ComfyUI 0.22.0. What’s Changed Add test that each model has unique identifiers (CORE-134) by @Talmaj in https://github.com/Comfy-Org/ComfyUI/pull/13654 Make audio processing nodes handle None -inputs by @kijai in https://github.com/C…
• Hugging Face Transformers 5.9.0. Release v5.9.0 New Model additions Cohere2Moe Command A+ is a Mixture-of-Experts (MoE) language model from Cohere that features a hybrid attention pattern combining sliding window and full attention layers.

Adds Cohere2Moe Command A+ model support.

• text-generation-webui 4.9. Changes - MTP speculative decoding support : Add draft-mtp as a new –spec-type option.

MTP speculative decoding now available via –spec-type draft-mtp.

• Claude Agent Sdk Typescript 0.3.144. What’s changed Assistant messages and StopFailure hooks now report error: ‘model not found’ when the selected model doesn’t exist or isn’t available, instead of the generic ‘invalid request’ .

Assistant messages now report model not found errors explicitly instead of generic invalid request.

• Rslearn 0.1.7. What’s Changed Remove incorrect GlobalPool/in channels additions; document PoolingDecoder for patch regression by @robmarkcole in 647 Add explicit support for v1 1 models in the EMBEDDING SIZES dict by @gabrieltseng i…

Read this brief on the web: https://freedomtech.news/posts/2026-05-21-bitcoin-daily-brief/