v2026.9.21
NousResearch/hermes-agentv2026.9.21Sep 21, 2026by teknium1
Upgrading?
AI Summary
This is a patch release (v0.21.4) aggregating approximately 1,800 merged PRs since v0.21.3. It provides a stable tagged release for downstream consumers like Docker images and Hermes Cloud, though full documentation is deferred to v0.22.0.
Key Highlights
- Aggregation of ~1,800 PRs and 5,000+ commits since the previous version.
- Desktop enhancements including attachment to the host backend, font picker, and one-click engine updates.
- CLI improvements featuring structured JSONL output via `--format stream-json`.
- Expansion of the video catalog with LTX 2.5 and Kling O3.
- Introduction of website pages for catalog plugins and authors.
New Features
- Host-wide gateway singleton lock with rendezvous record.
- Desktop attaches to the running host backend instead of spawning a second one.
- `--format stream-json` structured JSONL output for the CLI.
- `skills.auto_load` pinning skills into every new session's prompt.
- Desktop chat/UI font picker and one-click local engine updates.
- Plugin uninstall from the Plugins hub.
- `decline` unauthorized-DM behavior for the gateway.
- Configurable MCP discovery connect cap (`mcp.discovery_concurrency`).
- `session_search` with bounds and OR-relaxed recall retry.
- `hermes sessions set-journal-mode` command.
- LTX 2.5 and Kling O3 in the video catalogs.
- Website pages for every catalog plugin and author.
- Dozen new community plugins (tailscale, ssh, shodan, terminal, rss, resetwatch, done-bell, kiwi, cognee, Octen).
Full Release Notes
# Hermes Agent v0.21.4 (v2026.9.21) **Release Date:** September 21, 2026 > Patch release. This tag rolls up the ~1,800 PRs merged since v0.21.3 into a stable tagged release for downstream consumers (Docker images, Hermes Cloud, hosted deployments). Full curated notes for this window are deferred to v0.22.0. ## About this release Measured at commit `4b8a8134009a8727a289bcabeb0019fedd353128`, the window since v0.21.3 contains **5,071 non-merge commits** across **5,169 changed files** (+312,961 / −62,855), **1,812 merged PRs** and **2,116 closed issues**. Also in the window, undocumented here on purpose: a host-wide gateway singleton lock with a rendezvous record, and Desktop attaching to the running host backend instead of spawning a second one; one backend-owned connector operation with a setup card on Desktop, TUI and CLI; `--format stream-json` structured JSONL output for the CLI; `skills.auto_load` pinning skills into every new session's prompt; a Desktop chat/UI font picker, one-click local engine updates, and plugin uninstall from the Plugins hub; a `decline` unauthorized-DM behavior for the gateway; a configurable MCP discovery connect cap (`mcp.discovery_concurrency`); `session_search` after/before bounds plus OR-relaxed recall retry; `hermes sessions set-journal-mode`; LTX 2.5 and Kling O3 in the video catalogs; a website page for every catalog plugin and author, with pinned-commit READMEs and added/updated sorting; a dozen new community plugins in the catalog (tailscale, ssh, shodan, terminal, rss, resetwatch, done-bell, kiwi, cognee, Octen); and a large run of profile/multiplex isolation, cron, kanban, Desktop and state.db fixes. **Full curated release notes for this window ship with v0.22.0**, which will document everything from v0.21.0 onward — highlights, feature areas, and complete contributor credits. Nothing in this window is skipped. ## Updating - `hermes update` (git installs), or re-run the installer one-liner. - Docker / Hermes Cloud: images build from this tag (`nousresearch/hermes-agent:v2026.9.21`). **Full changelog:** https://github.com/NousResearch/hermes-agent/compare/v2026.9.14...v2026.9.21
Related Repositories
A collective list of free APIs
A feature-rich command-line audio/video downloader
Python tool for converting files and office documents to Markdown.
AutoGPT is the vision of accessible AI for everyone, to use and to build on. Our mission is to provide the tools, so that you can focus on what matters.
Build Agentic workflows, RAG pipelines, with rich AI model and tool support on one collaborative workspace. Deploy on cloud, VPC, or self-hosted, so teams move from prototype to production without rebuilding the stack.