Skip to content

feat(kvarn): per-token flash-attention kernels for q3_kvarn and q4_kvarn #1

feat(kvarn): per-token flash-attention kernels for q3_kvarn and q4_kvarn

feat(kvarn): per-token flash-attention kernels for q3_kvarn and q4_kvarn #1

Job Run time
1d 0h 0m 0s
1d 0h 0m 0s