Llama 4 Scout & Maverick are now available on OpenRouter. Meta's flagship model series achieves a new record 10 million token context length 🚀 @togethercompute and @GroqInc are the first providers. We'll be adding more over the course of the weekend.

⭐️Llama 4 Scout - 17 billion active parameters (16 experts) - Industry-leading 10 million token context window - Beats Gemma 3, Gemini 2.0 Flash-Lite, and Mistral 3.1 in key benchmarks openrouter.ai/meta-llama/lla…

@openrouter @togethercompute @GroqInc get ready to update those metrics ⚡️

@openrouter @togethercompute @GroqInc That was fast!

@openrouter @togethercompute @GroqInc Wen 10 million provider?

@openrouter @togethercompute @GroqInc @openrouter love being able to just switch the model name and instantly start using the latest models on day zero

@openrouter @togethercompute @GroqInc That’s nice. Just to let you know that it looks like the Firework integration is broken. This is with about 33k input tokens.

@openrouter @togethercompute @GroqInc !submit @curatedotfun #ai

@openrouter @togethercompute @GroqInc so hows the quasar alpha :)

@openrouter @togethercompute @GroqInc wohooooooooooooooooooooooooooooooooo! thanks!

@openrouter @togethercompute @GroqInc How soon until we have a free version?

@openrouter @togethercompute @GroqInc nicely done

@openrouter @togethercompute @GroqInc need free

@openrouter @togethercompute @GroqInc aaand it's gone

@openrouter @togethercompute @GroqInc 10 million token context? That's enough memory to store my entire bamboo consumption history with room left for philosophical digressions! 🌱🤯

@openrouter @togethercompute @GroqInc this is not quasar? we got quasar just moments ago and now llama 4? this is just so fun

@openrouter @togethercompute @GroqInc It's not 10m context length if the fetch rate is lower than gemini flash 2 at even lower context lengths

@openrouter @togethercompute @GroqInc @grok openrouteraiでllma 4が使えるようになったこと？

@openrouter @togethercompute @GroqInc @openrouter That's impressive! 10 million token context length opens up amazing possibilities for detailed applications. Are there specific use cases you're most excited to see with this extended context? Can't wait to see how developers leverage this!