llm, continue dropping keep alive since tools which use LLMs correctly manage this themselves.

2026-10-17 12:01:57 -07:00
parent 1e8aa6956f
commit 8a2a723d4b
1 changed files with 1 additions and 1 deletions
@@ -256,7 +256,7 @@
              export OLLAMA_REQUEST_TIMEOUT=3600 # default 300 (unconfirmed, not in the list of --help vars)

              ### Resource Usage ###
-              export OLLAMA_KEEP_ALIVE=2h # default is 5m
+              export OLLAMA_KEEP_ALIVE=20m # default is 5m
              export OLLAMA_MAX_LOADED_MODELS={{ llm_concurrency }} # default is 3 if using CPUs, or 3 * NumGPUs
              export OLLAMA_NUM_PARALLEL={{ llm_parallel }} # default is 1 and applies per-model (unconfirmed)
              export OLLAMA_CONTEXT_LENGTH={{ llm_context }} # default is 4096, https://docs.ollama.com/faq#how-can-i-specify-the-context-window-size