Gluon
A chip kernel programming language developed in-house by OpenAI, built on top of Triton. It adopts the SPMD programming model and provides a Linear Layout abstraction, supporting PTX-level primitives for low-level AI chip kernel development.
Timeline (last 90 days)
Gluon is a kernel programming language developed in-house by OpenAI, built on top of Triton, retaining the SPMD programming model while exposing lower-level programming abstractions
Gluon provides APIs for NVIDIA GPUs that can be mapped to PTX instructions, including MMA instructions, TMA instructions, and the M-Barrier mechanism
All Facts (2)
Gluon is a kernel programming language developed in-house by OpenAI, built on top of Triton, retaining the SPMD programming model while exposing lower-level programming abstractions
50%UnverifiedGluon provides APIs for NVIDIA GPUs that can be mapped to PTX instructions, including MMA instructions, TMA instructions, and the M-Barrier mechanism
50%