🚀 Introducing DeepInfer, our in-house inference framework, now fully open source! Powering DeepSeek-V4.1-Flash at scale with high throughput and low latency. We’re sharing our production stack to help the community lower compute costs, reduce energy use, and make AI more affordable. 🔗 github.com/deepseek-ai/DeepI… 1/4

🚀 Introducing DeepSeek-V4.1-Flash: smarter, faster, more efficient. 🔹 Introducing the smallest model in our new architecture family, with native visual understanding. 🔹 Designed for greater capability, faster inference, higher throughput, and scaling to larger models. 1/6

@thdxr I don’t understand the logic here. We get $60 of usage with the $10 OpenCode plan for DeepSeek. I entirely used the DeepSeek 4.1 Flash model, with only $3–4 of usage on the Muse Spark model. But with $44 of usage, how is my monthly limit already at 95%?

Claude Opus 5.5 made this video entirely with code.

Meta Muse Spark 1.3 is now free on OpenCode