LMSYS Org, the nonprofit behind the open-source inference engine SGLang, has launched the SGLang Consortium with eight founding members: Nvidia, AMD, Intel, Google, Meta, Cloudflare, Modal and RadixArk. The announcement came in the opening keynote of the inaugural SGLang Summit in San Francisco on 12 November.
The consortium will work with SGLang's maintainers to help shape the project's roadmap and advance open-source AI inference. It gives members a standing forum to discuss emerging inference trends and turn shared priorities into joint engineering work, focused on frontier model support, reasoning and multimodal workloads, distributed inference and hardware optimization.
Each founding member will designate a technical representative and join two roadmap meetings a year, contributing engineering expertise, deployment feedback, testing resources or joint development. Working groups may collaborate between meetings, with public updates on roadmap direction. Crucially, SGLang's maintainers retain responsibility for technical decisions and code acceptance, and the project stays open to the entire community.
The timing reflects inference's growing share of AI spending. According to Deloitte figures cited by LMSYS, inference is expected to account for roughly two-thirds of all AI compute in 2026, up from about half in 2025 — and the inference engine determines how fast and cost-effectively that compute runs on each company's hardware. SGLang, a full-stack inference engine in the PyTorch ecosystem, runs on hundreds of thousands of GPUs and powers large production deployments for companies including Meta, Microsoft, Google, Amazon, Thinking Machines Lab, Cloudflare, Reflection and Modal.
'Better software helps developers get more performance from NVIDIA compute,' said Ian Buck, Nvidia's vice president of hyperscale and high-performance computing. AMD's Ramine Roane said SGLang 'gives developers a fast, open inference engine that runs across hardware.' Ying Sheng, co-author of the original SGLang paper and CEO of RadixArk, said software approaches to efficiency will matter enormously given the difficulty of building new data centres.
In a related move, Intel announced a collaboration with RadixArk to optimize SGLang for Intel CPUs, GPUs and accelerators. The consortium is not exclusive to any vendor, and LMSYS stressed that the project remains open to contributors outside the founding group.

