Commit Graph
3 Commits
Author SHA1 Message Date
antirez bce69b05bb cuda: warm up top-k regression timing 2026-06-02 09:07:28 +02:00
user 2791d27eb3 cuda: support batched q4k routed moe 2026-05-30 10:22:01 +02:00
antirez 0c41d58c0c Add CUDA long-context regression test 2026-05-12 10:23:58 +02:00