You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@tooling.apache.org by ak...@apache.org on 2026/09/09 20:09:21 UTC
(tooling-llmao) 04/04: Adding selfhost api key to config example
This is an automated email from the ASF dual-hosted git repository.
andrewmusselman pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/tooling-llmao.git
commit 6ef0c657f30ad616d7bbe25441589305be6c4373
Author: Andrew Musselman <ak...@apache.org>
AuthorDate: Wed Sep 9 12:53:19 2026 -0700
Adding selfhost api key to config example
---
config.yaml.example | 8 ++++++++
1 file changed, 8 insertions(+)
diff --git a/config.yaml.example b/config.yaml.example
index 4e49bd4..d53b5d3 100644
--- a/config.yaml.example
+++ b/config.yaml.example
@@ -45,6 +45,14 @@ models_path: model_list.yaml
# Placement is keyed by the box public IP (request.remote_addr / X-Forwarded-For).
fleet:
key: CHANGE_ME_FLEET_KEY
+ # Bearer token every self-hosted vLLM is started with (--api-key) and that
+ # LiteLLM presents when calling it. One shared value for the fleet; a
+ # catalog entry may override per model, and per-instance keys generated at
+ # host-add time will eventually replace both.
+ #
+ # If this is unset AND the catalog carries no api_key, vLLM starts
+ # UNAUTHENTICATED on a public port.
+ selfhost_api_key: CHANGE_ME_SELFHOST_API_KEY
# Seconds between vLLM GET /health probes (unauthenticated; private path only).
health_interval_s: 45
# Per-probe HTTP timeout for vLLM /health.
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@tooling.apache.org
For additional commands, e-mail: commits-help@tooling.apache.org