From a857aa634c401026f80013ed68ea2e365179f683 Mon Sep 17 00:00:00 2001 From: Saylor Berman Date: Mon, 22 Sep 2025 14:43:38 -0600 Subject: [PATCH 1/2] Update diagram in inference doc The NJS model extraction module isn't needed at this time, so the diagram needs to be updated. --- docs/proposals/gateway-inference-extension.md | 16 +++++++--------- 1 file changed, 7 insertions(+), 9 deletions(-) diff --git a/docs/proposals/gateway-inference-extension.md b/docs/proposals/gateway-inference-extension.md index fb6abe3ae2..948f6506aa 100644 --- a/docs/proposals/gateway-inference-extension.md +++ b/docs/proposals/gateway-inference-extension.md @@ -76,20 +76,18 @@ flowchart TD A[Client Request] --> B[NGINX] subgraph NGINX Pod subgraph NGINX Container - B --1--> C[NJS Module: extract model name if needed] - C --2--> B - B --3--> D[NJS Module: Subrequest to Go App] + B --1--> D[NJS Module: Subrequest to Go App] end subgraph Go Application Container E[Go Application] end - D -- 4. subrequest --> E + D -- 2. subrequest --> E end - E -- 5. gRPC ext_proc protocol --> F[Endpoint Picker Pod] - F -- 6. Endpoint in Header --> E - E --7--> D - D --8--> B - B --9--> G[AI Workload Endpoint] + E -- 3. gRPC ext_proc protocol --> F[Endpoint Picker Pod] + F -- 4. Endpoint in Header --> E + E --5--> D + D --6--> B + B --7--> G[AI Workload Endpoint] ``` ## API, Customer Driven Interfaces, and User Experience From 1020968f6c41ebbf7dc63556c4eb06b5fa50d603 Mon Sep 17 00:00:00 2001 From: Saylor Berman Date: Mon, 22 Sep 2025 15:03:00 -0600 Subject: [PATCH 2/2] Fix lettering --- docs/proposals/gateway-inference-extension.md | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/docs/proposals/gateway-inference-extension.md b/docs/proposals/gateway-inference-extension.md index 948f6506aa..bf17bf51d4 100644 --- a/docs/proposals/gateway-inference-extension.md +++ b/docs/proposals/gateway-inference-extension.md @@ -76,18 +76,18 @@ flowchart TD A[Client Request] --> B[NGINX] subgraph NGINX Pod subgraph NGINX Container - B --1--> D[NJS Module: Subrequest to Go App] + B --1--> C[NJS Module: Subrequest to Go App] end subgraph Go Application Container - E[Go Application] + D[Go Application] end - D -- 2. subrequest --> E + C -- 2. subrequest --> D end - E -- 3. gRPC ext_proc protocol --> F[Endpoint Picker Pod] - F -- 4. Endpoint in Header --> E - E --5--> D - D --6--> B - B --7--> G[AI Workload Endpoint] + D -- 3. gRPC ext_proc protocol --> E[Endpoint Picker Pod] + E -- 4. Endpoint in Header --> D + D --5--> C + C --6--> B + B --7--> F[AI Workload Endpoint] ``` ## API, Customer Driven Interfaces, and User Experience