OpenClaw v2026.4.5

v2026.4.5 dropped yesterday with some major features. Video generation, music generation, bundled ComfyUI support, new providers, and a multilingual Control UI. Here's the highlights.

Video Generation

Built-in video_generate tool lets agents create videos through configured providers and return the generated media directly in the reply. Bundled providers include xAI (grok-imagine-video), Alibaba Model Studio Wan, and Runway.

Music Generation

New music_generate tool with bundled Google Lyria and MiniMax providers plus workflow-backed Comfy support. Includes async task tracking and follow-up delivery of finished audio.

ComfyUI Integration

Bundled ComfyUI workflow media plugin for local ComfyUI and Comfy Cloud workflows. Includes shared image_generate, video_generate, and workflow-backed music_generate support with prompt injection, optional reference-image upload, live tests, and output download.

New Providers

• Qwen — bundled provider
• Fireworks AI — bundled provider
• StepFun — bundled provider
• MiniMax TTS — speech integration
• Amazon Bedrock Mantle — inference profile discovery
• Ollama Web Search — search integration
• MiniMax Search — chat and search workflows

Multilingual Control UI

Control UI now supports: Simplified Chinese, Traditional Chinese, Brazilian Portuguese, German, Spanish, Japanese, Korean, French, Turkish, Indonesian, Polish, and Ukrainian.

Memory & Dreaming

Experimental features continue to ship: weighted short-term recall promotion, /dreaming command, Dream Diary surface, multilingual conceptual tagging, and configurable aging controls.

Breaking Changes

Legacy public config aliases removed (talk.voiceId, talk.apiKey, agents.*.sandbox.perSession, browser.ssrfPolicy.allowPrivateNetwork, hooks.internal.handlers, channel/group/room toggles). Use canonical paths with openclaw doctor --fix migration support for existing configs.