Sticky routing for whole conversations
Sessions
Group related requests with an x-session-id header to keep provider routing sticky and trace a conversation end to end.
x-session-id → Sticky routing
Same provider, warm prompt cache
01
One header
Add x-session-id. Nothing else.
02
Warm caches
Stay on the provider holding your cache.
03
On by default
Stickiness works out of the box.
04
Session view
Follow a conversation in the activity log.
Built for
Coding agents
Long tool loops stay cache-friendly
Chat apps
Consistent behavior across turns
Debugging
Follow one user's whole conversation
Tag a conversation
terminal
curl https://api.llmgateway.io/v1/chat/completions \ -H "Authorization: Bearer $LLM_GATEWAY_API_KEY" \ -H "Content-Type: application/json" \ -H "x-session-id: conversation-9f8e7d6c" \ -d '{ "model": "claude-sonnet-4-6", "messages": [ { "role": "user", "content": "Hello!" } ] }'