Have been playing around with @OpenRouter 's new Auto Beta, and I have to say it's awesome. Really tempted to ship this Auto-router to Prod on Shiori. It's really good at selecting the right model for the task at hand. (Used DeepSeek-V4-Pro in this pic)

Launched today on OpenRouter: Gemini 3.6 Flash and Gemini 3.5 Flash-Lite! Both strong updates to their model families, with high throughput (150+ tok/s) for agents, from token-efficient coding and knowledge work to low-latency, high-volume subagents. Details below 🧵

Gemini 3.6 Flash: improved coding, knowledge work, multimodal performance, and agentic planning. Google reports 17% fewer output tokens than 3.5 Flash. openrouter.ai/google/gemini-… Link Gemini 3.6 Flash - API Pricing & Providers Gemini 3.6 Flash is a high-efficiency model…

Gemini 3.5 Flash-Lite: Google’s fastest 3.5-series model, built for low-latency, high-throughput agentic search, document processing, and subagents. Running on OpenRouter at 150+ tok/s openrouter.ai/google/gemini-… Link Gemini 3.5 Flash-Lite - API Pricing &…

Laguna S 2.1 from @poolsideai is live on OpenRouter! An open-weight 118B-A8B MoE for agentic coding and long-horizon work, with a 1M context window. Scores: 70.2% on Terminal-Bench 2.1 and 40.4% on DeepSWE. Try it: openrouter.ai/poolside/lagun… Link Laguna S 2.1 - API…