Red Hat AI Releases vLLM Checkpoints to Accelerate Inference

vllm_project · x · 2026-07-16

Red Hat AI has released DFlash speculator checkpoints for two open-source NVIDIA LLMs:

Trained on vLLM's open-source Speculators library under the Apache 2.0 license, they are validated on NVIDIA B200. The official vLLM activation parameter is: --spec-model ... --spec-tokens 7 --spec-method dflash.

Performance-wise, math and reasoning tasks accept an average of 5 out of roughly 7 draft tokens, while coding tasks (HumanEval) accept an average of 3.4 out of 7.

Related event: Red Hat AI Releases DFlash Inference Checkpoints for NVIDIA Models(2 posts)→

Original post →

More from Infra

Infra channel →