Gemini 3.6 Flash and Gemini 3.5 Flash-Lite are live in Command Code. Available on Pro, Max and Team plans Gemini 3.6 Flash: 🔹 Better coding, knowledge work and multimodal performance 🔹 17% fewer output tokens vs 3.5 Flash, with a lower cost per token Gemini 3.5 Flash-Lite: 🔹 Fast, cost effective 3.5-class model, ideal for subagents 🔹 350 output tokens/second Try now with $ cmd --model google/gemini-3.6-flash $ cmd --model google/gemini-3.5-flash-lite

We support both frontier open and commercial models from all the leading AI Labs. To learn more about these, read the docs here: commandcode.ai/docs/reference…

@CommandCodeAI Laguna where?

@CommandCodeAI All these numbers are crazy. 350 tokens output/sec??? Can't wait to try it out.

@CommandCodeAI Love seeing faster and more cost-efficient AI models. Lower latency and lower costs make it easier to build AI products at scale. 🚀

@CommandCodeAI The token efficiency on 3.6 Flash is honestly underrated. Fewer output tokens while improving quality is a nice win also 350 output tokens/sec on Flash-Lite is kinda wild

@CommandCodeAI Let's go! 👊

@CommandCodeAI Eu estourei meu limite todo semanal e quase mensal com alguns comandos usando o Kimi K3 pra testar. Agora só consigo fazer algo mês que vem 🤣🤣

@CommandCodeAI 🔷🔷🔷

@CommandCodeAI 3.6 Flash very dump, it's not smart than 3.5 flash