b6220

sched : copy only the used experts when offloading prompt processing …

b6209

opencl: mark `argsort` unsupported if cols exceed workgroup limit (#1…

b6170

CUDA: fix negative KV_max values in FA (#15321)

b6153

perplexity: give more information about constraints on failure (#15303)

* perplexity: give more information about constraints on failure

This checks whether -np is insufficient vs context, and provides clues as to how much is needed for each.

* log formatting

* log error and return instead of storing max_seq_exceeded int

* check if s0 is zero for -np check

b6144

ggml : repack block_iq4_nlx8 (#14904)

ggml-ci