releaseintermediate
[Release] openclaw/openclaw v2026.4.24-beta.3: openclaw 2026.4.24-beta.3
By steipetegithub
View original on githubOpenClaw v2026.4.24-beta.3 introduces Google Meet as a bundled participant plugin with real-time voice capabilities, adds DeepSeek V4 models to the default catalog, and enhances browser automation with coordinate clicks and improved tab recovery. The release includes significant infrastructure improvements for lighter startup times through static model catalogs and lazy provider dependencies, plus numerous bug fixes across Telegram, browser automation, MCP, sessions, and gateway restart handling.
Key Points
- •Google Meet integration: bundled plugin with personal Google auth, Chrome/Twilio realtime sessions, artifact/attendance exports, and recovery tooling for existing Meet tabs
- •DeepSeek V4 Flash and Pro models now in bundled catalog with V4 Flash as onboarding default; fixed thinking/replay behavior for follow-up tool-call turns
- •Real-time voice loops for Talk, Voice Call, and Google Meet enable deeper tool-backed answers by consulting full OpenClaw agent
- •Browser automation improvements: coordinate clicks, longer default action budgets, per-profile headless overrides, and steadier tab reuse/recovery
- •Lighter startup infrastructure: static model catalogs, manifest-backed model rows, lazy provider dependencies, and external runtime-dependency repair
- •Fixed packaged installs to preserve package-root runtime dependencies when bundled plugin runtime mirrors fall back to copying shared chunks
- •Heartbeat scheduler fix: clamp oversized delays through safe timer helper to prevent Node timeout cap crashes
- •Telegram improvements: removed startup preflight conflicts, webhook acknowledgment before middleware execution, and duplicate long-poller prevention
- •Browser/Playwright fixes: ignore benign route races during navigation, bind ARIA snapshot refs to live DOM nodes, and improved Chromium detection on Linux
- •Session management: close idle browser tabs on rollover, fall back to transcript-estimated token counts for oversized thread forks, and durable restart continuation handling
Found this useful? Add it to a playbook for a step-by-step implementation guide.
Workflow Diagram
Start Process
Step A
Step B
Step C
Complete