Increase llama prompt cache to 32 GiB

2026-09-03 10:03:43 +02:00
parent 4153e535d3
commit fb0cb40bed
4 changed files with 8 additions and 8 deletions
@@ -49,7 +49,7 @@ UNCENSORED_MTP_MAX=2
 LLAMA_THREADS=6
 LLAMA_THREADS_BATCH=6
 FAST_PARALLEL_SLOTS=1
-LLAMA_CACHE_RAM_MIB=24576
+LLAMA_CACHE_RAM_MIB=32768
 MEDIUM_PARALLEL_SLOTS=1
 LARGE_PARALLEL_SLOTS=1
 ULTRA_PARALLEL_SLOTS=1