Entities · Products
SGLang
4 articles tagged with this entity.
-
Capturing token IDs during agentic interactions for better reinforcement learning
-
BlendServe: Optimizing Offline Inference for Auto-regressive Large Models with Resource-aware Batching
-
FastKernels: Benchmarking GPU Kernel Generation in Production
-
Making Every Verified Token Count: Adaptive Verification for MoE Speculative Decoding