From d40a0946e198bf6f322dd874435fb947285600ee Mon Sep 17 00:00:00 2001 From: Admin Bot Date: Tue, 8 Sep 2026 18:26:32 -0700 Subject: [PATCH] feat: route qwen2.5:3b-instruct to CPU service on cp-2 Model moved from ornith-predictor GPU to dedicated qwen-cpu Deployment on talos-cp-2 (CPU). Frees 1 GPU for ComfyUI. Co-authored-by: poimen --- k8s/configmap.yaml | 2 +- k8s/gateway-config-secret.enc.yaml | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/k8s/configmap.yaml b/k8s/configmap.yaml index d04904f..f5a0807 100644 --- a/k8s/configmap.yaml +++ b/k8s/configmap.yaml @@ -36,7 +36,7 @@ data: path: "/v1/chat/completions" - name: "qwen2.5:3b-instruct" - address: "ornith-predictor.llm-serving:80" + address: "qwen-cpu.llm-serving:80" path: "/v1/chat/completions" - name: "nomic-ai/nomic-embed-text-v2-moe" diff --git a/k8s/gateway-config-secret.enc.yaml b/k8s/gateway-config-secret.enc.yaml index a02cf3e..beaee53 100644 --- a/k8s/gateway-config-secret.enc.yaml +++ b/k8s/gateway-config-secret.enc.yaml @@ -25,7 +25,7 @@ stringData: address: "ornith-predictor.llm-serving:80" path: "/v1/chat/completions" - name: "qwen2.5:3b-instruct" - address: "ornith-predictor.llm-serving:80" + address: "qwen-cpu.llm-serving:80" path: "/v1/chat/completions" - name: "nomic-ai/nomic-embed-text-v2-moe" address: "embeddings-predictor.llm-serving:80"