Simple Application Server:OpenClaw application image release notes

Released on May 29, 2026

Released the OpenClaw 2026.5.19 image. We recommend resetting your system to the latest image to get the following enhancements.

• Upgraded the built-in software to OpenClaw 2026.5.19.

• You can now configure the Alibaba Cloud Model Studio Token Plan (Team Edition) and multiple third-party model platforms on the application details page.

• Added support for configuring bots by scanning a QQ QR code.

• Added support for displaying installed skills from the workspace directory.

• Switched to the Alibaba Cloud Linux 4 LTS 64-bit Agentic Edition operating system.

Released on April 17, 2026

Released the OpenClaw 2026.4.14 image. We recommend resetting your system to the latest image to get the following enhancements.

• Upgraded the built-in software to OpenClaw 2026.4.14.

• The application details page now supports the qwen3.6-plus model. You can switch to this model directly in the console.

Released on April 2, 2026

This image upgrades the built-in software to OpenClaw 2026.3.28.

Released on March 26, 2026

This image upgrades the built-in software to version 2026.3.13.

Updates since the OpenClaw 2026.3.3 application image:

• Upgraded official plugins: The official plugins for DingTalk, Lark, and WeCom were upgraded to improve channel stability and functionality.

• Optimized WeCom channel: The WeCom channel now uses a persistent connection instead of URL callbacks, improving message response speed and stability.

• Improved configuration experience: Added a model configuration validation feature to prevent issues caused by incorrect model settings.

Released on March 5, 2026

Released the OpenClaw 2026.3.3 image, which upgrades the built-in software to OpenClaw 2026.3.3 and incorporates the latest features from the community.

Updates since the OpenClaw 2026.2.26 application image:

• Built-in headless browser configuration: Pre-configured a managed Chrome headless browser, allowing OpenClaw to use dedicated user profiles for more secure and stable browser operations.

• Pre-installed core enhanced skills: This release pre-configures several fundamental skills to improve the AI assistant's intelligence and automation. These include:

  • find-skills: Discovers and installs useful skills.

  • skill-vetter: Performs security scans on skills to ensure a safe runtime environment.

  • agent-browser: Enables browser automation.

  • self-improving-agent: Enables the AI assistant to learn from its mistakes and improve over time.

  • proactive-agent: Enhances the proactivity of OpenClaw.

• Enhanced DingTalk channel: You can now send attachments such as images and compressed files (up to 5 MB) to OpenClaw through direct messages in DingTalk.

• Fixed Lark channel issue: This update optimizes account configuration for the Lark channel, resolving an issue where scheduled tasks failed to execute for some users.

Released on February 28, 2026

Released the OpenClaw 2026.2.26 image, which upgrades the built-in software to OpenClaw 2026.2.26 and incorporates the latest features from the community.

Updates since the OpenClaw 2026.2.9 application image:

• Native multimodal model support: Pre-configured the qwen3.5-plus native multimodal model. Users can now process both text and image inputs by using this single model.

• Memory and stability optimization: Pre-configured 2 GiB of swap virtual memory to mitigate issues caused by out-of-memory errors. (Enabling a swap partition may decrease memory I/O performance. If your instance runs low on memory, we recommend upgrading your instance configuration to increase the physical memory.)

Updated on February 26, 2026

• Model and compatibility optimization. Improved compatibility with the latest Coding Plan. On the application details page in the Simple Application Server console, you can switch freely between models such as qwen3.5-plus, kimi-k2.5, glm-5, and MiniMax-M2.5.

• Added multimodal capability support. OpenClaw now supports image inputs from the qwen3.5-plus and kimi-k2.5 models. You can now process both text and image inputs by using a single model.

Released on February 11, 2026

Released OpenClaw 2026.2.9, which incorporates community features as of February 9, 2026.

Updates since the OpenClaw 2026.2.3 application image:

• Fixed an issue with web search. SearXNG is now prioritized for web searches.

• In addition to AppFlow bridge mode, this release includes built-in plugins for DingTalk, Lark, WeCom, and QQ. You can configure these services by entering their keys on the application details page.

• Added a personalization panel for the AI assistant. This panel includes a built-in Markdown editor to directly modify the following files: SOUL.md, IDENTITY.md, and AGENTS.md.

• You can now modify models in the console in the following ways:

  1. Switch between different Model Studio models directly from a drop-down menu. This makes it easier to change models without editing the OpenClaw configuration file.

  2. You can also manually enter specific models supported by Model Studio.

• The initialization process now generates a random service port by default. This effectively hides the service endpoint and helps prevent malicious scans.

• Public network access control: You can now enable or disable public network access with a single click.

• Gateway restart: You can now restart the gateway service from the user interface.

Released on February 5, 2026

Released the OpenClaw 2026.2.3 version, which upgrades the built-in software and incorporates the latest features from the community as of February 3, 2026.

What's new compared to the OpenClaw (Moltbot) 2026.1.27-beta.1 release?

• Web search, using the built-in SearXNG web search skill, is now available by default in all regions. No extra configuration or fees are required.

• Supports synergistic text processing and image recognition, where the Image Model for image recognition defaults to the qwen3-vl-plus model.

• Docker is now built into OpenClaw 2026.2.3. This allows agents to use a Docker runtime as a sandbox.