my team is always working through a phat backlog of inference providers that want to onboard onto openrouter
right now i’m prioritizing media gen (image gen, audio, video)
what else do people want to see?
Inkling from @thinkymachines is live on OpenRouter.
An open-weights MoE model with 975B total / 41B active parameters, 1M context, and controllable reasoning across text, images, and audio.
Try it:
server tools are an underrated feature on @OpenRouter
here's how to use the Advisor and Subagent server tools to get more leverage out of frontier models like Fable and GPT 5.6 by pairing them with cheaper models
At @sentry we just started to adopt @OpenRouter so I quickly wiped together a @pidotdev extension that exposes their most useful server tools as Pi tools. Took ~15mins to have this for idea to finished repo, Pi is king
New and exciting model family on @OpenRouter: Muse Spark, from Meta!
We want to make sure that many choices for AGI surface in the market, not just one.
Things to try:
1. Ask the OpenRouter MCP where in your codebase you should try it
2. Use it in the chatroom:
Developer choice is core to what we're building.
We’re excited to share that Muse Spark 1.1 is now available on @OpenRouter for US-based developers.
Get started today 👉 bit.ly/4vvjSNa
Muse Spark 1.1 from @AIatMeta is live on OpenRouter for US-based devs!
At $1.25/M in and $4.25/M out, the model is poised to be one of the most price-efficient, high-intelligence models on OpenRouter. Built for production agentic workloads w/ multimodal inputs, viable at scale.
Muse Spark 1.1 is a multimodal reasoning model designed for agentic tasks, including coding, tool use, computer use, and multimodal understanding. See how it performs across these capabilities:
One command. A working AI coding agent.
Stripe Projects Build scaffolds our Daytona Coding Agent template: an LLM writes and runs code in a @daytonaio sandbox, live-previewed in chat via @CopilotKit. Any model via @OpenRouter, deploy on @vercel.
Live now in Stripe Projects