OpenClaw 2026.10.1: Agent Platform with Local Gateway and Messenger Channels
OpenClaw 2026.10.1 is out — a release dated 10.10.2026. It is an open-source agent environment: the assistant runs on your computer and connects to Discord, iMessage, Slack, Teams, Telegram, WhatsApp and more than 20 other channels, as well as to native apps for macOS, iOS, Android, Windows and Linux.
What's new
Six releases have come out since the last review — from 2026.9.5 to 2026.10.1. The main changes:
- Sessions and memory — usage is preserved across registry changes, attachments are delivered from remote workspaces, embedding caches are migrated in bounded batches with a report on oversized rows.
- Replies and media — local video playback is restored, rejected media links are replaced with useful errors, voice-only replies with reasoning enabled are fixed.
- Updates and Doctor — serving-verdict recovery hints are improved, repair works with a read-only managed configuration, successful cleanup is shown as progress without corrupting JSON output.
- Windows and browser — creation of empty and nested work trees is fixed, Chromium detection on Linux is improved, Playwright Chromium launches automatically on ARM64.
- Cloud workers and Crabbox — the real cloud worker error is shown, Linux leases are enforced, worker bundle downloads overlap with bootstrap.
What the platform does
OpenClaw is a local Gateway that manages sessions, tools, events and channel connections. The Control UI, CLI and TUI connect to it. Tools, skills and plugins extend the assistant's capabilities, while companion apps and nodes add voice, Canvas, camera, screen and on-device actions.
Your own model instead of the cloud
You can connect your own model via Ollama, an OpenAI-compatible API or MLX. It installs via npm: npm install -g openclaw@latest --allow-scripts=openclaw. License MIT — any use, modification and distribution is permitted, including commercial, provided the copyright notice and license text are retained.
How it differs from its neighbors
According to a comparison with OpenBot, Hermes Agent and DeerFlow: OpenClaw claims its own tools and functions, as well as its own server and Docker — just like all its neighbors. MCP support, agent memory and multiple agents working together are not mentioned in OpenClaw's documentation, whereas most of its neighbors claim them. For your own model, OpenClaw offers three connection methods — Ollama, an OpenAI-compatible API and MLX; OpenBot has only an OpenAI-compatible API, Hermes Agent has Ollama, llama.cpp and vLLM, and DeerFlow has an OpenAI-compatible API and vLLM.
| OpenClaw | OpenBot | Hermes Agent | DeerFlow | |
|---|---|---|---|---|
| Type | environment | environment | environment | environment |
| Language | TypeScript | TypeScript | Python | Python |
| License | MIT | MIT | MIT | MIT |
| GitHub stars | 391,715 | 5,112 | 246,743 | 82,638 |
| Releases in 90 days | 18 | 12 | 16 | 1 |
| Multiple agents working together | — | — | yes | yes |
| Own tools and functions | yes | yes | yes | yes |
| MCP support | — | yes | yes | yes |
| Agent memory | — | yes | yes | yes |
| Sandboxed code execution | — | — | — | yes |
| Browser interaction | — | yes | — | — |
| Human approval | — | — | yes | — |
| Tracing and debugging | — | — | — | yes |
| Voice and speech | yes | — | yes | — |
| Own server and Docker | yes | yes | yes | yes |
| Own model: Ollama | yes | — | yes | — |
| Own model: llama.cpp | — | — | yes | — |
| Own model: vLLM | — | — | yes | yes |
| Own model: MLX | yes | — | — | — |
| Own model: OpenAI-compat. API | yes | yes | — | yes |
Comparison based on the projects' READMEs and documentation as of 12.10.2026: "yes" means the capability is claimed there, a dash means it is not mentioned (which does not mean the capability is absent). Stars and releases are per GitHub data.
Those who already use OpenClaw and have been waiting for fixes in session handling, media and the Windows environment should update. Those looking for an agent environment with local control and a broad set of channels should take a closer look. You can follow new releases in the CPU3D news feed.
