Remix.run Logo
Topfi 4 hours ago

Unless I misunderstood what they wrote, I read parallelized in the diffusion sense, akin to GemmaDiffusion and Inception Labs models. Incidentally, Mercury 2.5 is truly groundbreaking, giving it a try is highly recommended.

mohsen1 an hour ago | parent [-]

yup https://github.com/vllm-project/vllm/pull/57250