@measure_plan@threejs Awesome! If this tech could handle browser controls, that’d be incredible. Scrolling and clicking with voice and gestures would let me browse the web while doing the dishes 🧼✨
@FlowiseAI I tested n8n's new MCP support in v1.88.0. Works well with Claude 3.5 but has issues with Gemini 2.0 Flash. @FlowiseAI and @n8n_io both have their strengths! Looking forward to seeing how MCP support evolves across platforms. #MCP#LLM
@FlowiseAI I've spent this week searching for automation tools that properly run MCP. After comparing Dify, n8n, and Flowise, I found that Flowise works most accurately! It's particularly impressive how it runs stably even with weaker LLM models like Gemini 2.0 Flash.
@FlowiseAI I tested n8n's new MCP support in v1.88.0 (unofficial). Works fine with Claude 3.5 but errors with Gemini 2.0 Flash. @FlowiseAI still has the edge with stability on weaker models! Working across more LLM models is Flowise's big advantage.
We’re adding native support for the MCP protocol in n8n: you can now get early access to both our MCP Client tool & MCP Trigger nodes!
Try it out, break it, build something cool, and let us know what you need.
👉 https://t.co/2TnqGvVXmn
#n8n#AI#MCP