Tagged "parallel-speculative-decoding"

• P-EAGLE: Faster LLM Inference with Parallel Speculative Decoding in vLLM 14 March 2026