# LMSYS Org - **Website:** - **Primary alias:** `sglang.ai` ## Description SGLang is an innovative and rapidly evolving framework designed for the efficient serving of large language models and vision language models. This open-source project empowers developers with a powerful backend runtime that features advanced capabilities such as RadixAttention for caching, a jump-forward decoding system, and overhead-free scheduling solutions. SGLang's flexible frontend language provides an intuitive interface for creating sophisticated LLM applications, allowing for chained generation calls, multi-modal inputs, and fine-tuned control flow. The framework supports an extensive array of generative models, including prominent architectures like Llama and Mistral, along with embedding and reward models. With its growing user community and industry adoption, SGLang not only streamlines model interactions but also fosters collaboration among developers through resources such as tutorials and a development roadmap. Supported by a range of industry leaders, SGLang stands out as a pivotal tool for those looking to harness the power of AI and machine learning, making complex tasks more manageable and efficient. ## Industries - 🖥 Computers Electronics and Technology - 🖥 Artificial Intelligence and Machine Learning _(under Computers Electronics and Technology)_ ## Links - [twitter](https://twitter.com/lmsysorg) - [linkedin](https://linkedin.com/company/sgl-project) - [github](https://github.com/sgl-project) ## Logos & icons - logo _(primary)_ — PNG — [download](https://cdn.brandfetch.io/idKbs-qdZ_/w/2392/h/728/theme/dark/logo.png?c=1bxid64Mup7aczewSAYMX&t=1761167321071) - icon _(primary)_ — JPEG — [download](https://cdn.brandfetch.io/idKbs-qdZ_/w/400/h/400/theme/dark/icon.jpeg?c=1bxid64Mup7aczewSAYMX&t=1790128519950) ## Colors | Name | Hex | Theme | | --- | --- | --- | | Orange Roughy | #d55815 | accent | | Cod Gray | #181717 | dark | | Alabaster | #fafafa | light | ## Fonts - monospace — asset _(custom)_ --- _Source: _ _See [/llms.txt](/llms.txt) for a full list of agent-readable pages, and [/auth.md](/auth.md) for how agents authenticate._