3 Commits

Author	SHA1	Message	Date
OlivierDehaene	ab7ccf5bc3	feat: add payload limit (#2726) ... * feat: add payload limit * update launcher	2024-11-21 18:20:15 +00:00
Nicolas Patry	a5593ba83e	Hotfixing auto length (warmup max_s was wrong). (#2716)	2024-11-04 09:55:54 +01:00
OlivierDehaene	6f88bd9390	feat: add triton kernels to decrease latency of large batches (#2687) ... * feat: add triton kernels to decrease latency of large batches * cast to int32 * fix kernel * fix kernel * disable triton on rocm * fix speculation * add slots filtering kernel	2024-10-25 21:10:00 +00:00