OrcaRouter 🐳

Prompt-aware routing. 150+ models supported.
Zero markup. Open source. One API.
Join the builders →

[OrcaRouter 🐳 @OrcaRouter]

🎉Introducing OrcaRouter Monthly Plans

Get rewarded for committing to scale.

Starting today, monthly subscriptions receive up to 10% bonus credits automatically every billing cycle.

Use those credits across every model on OrcaRouter:
• Claude
• Gemini
• GPT
• Grok
• Open-source models like Deepseek, Qwen, GLM, Minimax, Kimi, and more

No lock-in. No model restrictions. Just more AI for the same spend. Start building → https://t.co/BUgRhJdrhg

[OrcaRouter 🐳 @OrcaRouter]

https://t.co/03IkvuYBV0

[OrcaRouter 🐳 @OrcaRouter]

🎉Introducing OrcaRouter Monthly Plans

Get rewarded for committing to scale.

Starting today, monthly subscriptions receive up to 10% bonus credits automatically every billing cycle.

Use those credits across every model on OrcaRouter:
• Claude
• Gemini
• GPT
• Grok
• Open-source models like Deepseek, Qwen, GLM, Minimax, Kimi, and more

No lock-in. No model restrictions. Just more AI for the same spend. Start building → https://t.co/BUgRhJdrhg

[OrcaRouter 🐳 @OrcaRouter]

@jp54362 Both

[OrcaRouter 🐳 @OrcaRouter]

@DanKulkov You can use us to route complex tasks to Claude and mundane tasks to Deepseek

[OrcaRouter 🐳 @OrcaRouter]

@jayair @opencode Congrats 🎉

[OrcaRouter 🐳 @OrcaRouter]

@faychuk @garrytan also OrcaRouter 🫡

[OrcaRouter 🐳 @OrcaRouter]

@sdusebout @garrytan The right model router still makes the difference

[OrcaRouter 🐳 @OrcaRouter]

OrcaRouter + @Ragflowai = Production-Ready AI Infrastructure.

Combine RAGFlow's advanced RAG & Agent framework with OrcaRouter. Use the best model for every retrieval, reasoning, and agent task — without changing your application code.

#RAGFlow #LLMOps #AIEngineering #Agents #RAG

[OrcaRouter 🐳 @OrcaRouter]

We dug in: on v4-flash, requests that have thinking turned on AND set tool_choice to "required" (or a specific function) are getting rejected upstream with a 400. An easy one to work around — with thinking on, just set tool_choice to "auto" (or leave it off); and if you need a forced tool call, send that request without thinking and you'll be all set.

[OrcaRouter 🐳 @OrcaRouter]

🚀 Qwen 3.7 Plus is now live on OrcaRouter. Access it instantly through our API: https://t.co/JM2omnOpI3

#Qwen #AI #OrcaRouter

[OrcaRouter 🐳 @OrcaRouter]

@rangga_dikta Might be rate limits etc, but you get auto fail-overed to over models if you use router

[OrcaRouter 🐳 @OrcaRouter]

@oneitonitram @teortaxesTex Try it out on OrcaRouter

[OrcaRouter 🐳 @OrcaRouter]

@andimarketing2 @MiniMax_AI just use it on OrcaRouter

[OrcaRouter 🐳 @OrcaRouter]

@toi500 It’s live btw

[OrcaRouter 🐳 @OrcaRouter]

We’ve been quietly building nonstop at OrcaRouter — and today we’re releasing the OrcaRouter MCP Server🚀

Connect Claude Desktop, Cursor, Windsurf, Zed, OpenClaw, and any MCP client to 200+ AI models through one unified MCP interface.

✨ Features:
🗺️ Discover providers & models without an API key
💬 Run chat completions through any served model
🧠 Adaptive auto-routing
🔁 Built-in fallback chains for resilience
📊 Server-side filtering by provider, capability, or context window
🎯 Detailed model cards with pricing, latency & supported endpoints

One MCP server. Every model. Zero vendor lock-in. https://t.co/898e7Qyixp

#MCP #AI #Claude #CursorAI #LLM #AgenticAI

[OrcaRouter 🐳 @OrcaRouter]

@MiniMax_AI This is 🔥 🔥 🔥

[OrcaRouter 🐳 @OrcaRouter]

@MiniMax_AI Its 50% off for the first 7 days through June 7th!

[OrcaRouter 🐳 @OrcaRouter]

🚀 @MiniMax_AI M3 is now available on OrcaRouter. One of the most anticipated open model releases, bringing next-gen sparse attention, ultra-long context handling (up to 1M tokens of context with a guaranteed minimum of 512K), faster inference, and stronger agentic workflows.

Try it out here: https://t.co/gAnSOjFsy8

[OrcaRouter 🐳 @OrcaRouter]

@KaiXCreator Disagree

[OrcaRouter 🐳 @OrcaRouter]

@bindureddy Why do you think so? Opus 4.8 seems to be stronger

[OrcaRouter 🐳 @OrcaRouter]

@JinjingLiang lol