How vLLM implements distribution-preserving Gumbel-max text watermarking with efficient GPU kernels, statistical detection, speculative decoding, and repeated-context safeguards.
Watermarking in vLLM
calendar_today
September 24, 2026
person
Raphaël Rialland (Mistral), Simon Veitner (Red Hat), and Tomas Ruiz (Red Hat)
domain
vllm