diff --git a/.changeset/llm-stack-fix-config-updates.md b/.changeset/llm-stack-fix-config-updates.md
new file mode 100644
index 00000000..3b287542
--- /dev/null
+++ b/.changeset/llm-stack-fix-config-updates.md
@@ -0,0 +1,5 @@
+---
+"llm-stack": patch
+---
+
+Made the digest header for the apisix config update variable so that config updates are accepted by apisix
\ No newline at end of file
diff --git a/charts/llm-stack/bin/test-request-scaleway.sh b/charts/llm-stack/bin/test-request-scaleway.sh
index 2815e882..da73ad40 100755
--- a/charts/llm-stack/bin/test-request-scaleway.sh
+++ b/charts/llm-stack/bin/test-request-scaleway.sh
@@ -1,7 +1,7 @@
 PORT=${PORT:-80}
 HOST=${HOST:-llm-stack.localhost}
 
-curl http://$HOST:$PORT/v1/chat/completions \
+curl http://$HOST:$PORT/v1/chat/completions -v \
   -H "Content-Type: application/json" \
   -H "Authorization: Bearer sk-client-v1-abcdef123456" \
   -d '{
diff --git a/charts/llm-stack/examples/scaleway.yaml b/charts/llm-stack/examples/scaleway.yaml
index d41ab69b..34a7434a 100644
--- a/charts/llm-stack/examples/scaleway.yaml
+++ b/charts/llm-stack/examples/scaleway.yaml
@@ -4,11 +4,6 @@ apisixInitialConfig:
     provider: openai-compatible
     # retrieve the project id from the scaleway project dashboard
     baseURL: "https://api.scaleway.ai/<project_id>/v1/chat/completions" # ignored if vllm is enabled
-    models:
-      - name: qwen3.6-35b-a3b
-        provider: scaleway
-      - name: gemma-4-26b-a4b-it
-        provider: scaleway
 
 apisix:
   metrics:
diff --git a/charts/llm-stack/templates/apisix-config-seeder-configmap.yaml b/charts/llm-stack/templates/apisix-config-seeder-configmap.yaml
index d7b06f09..1ee21a33 100644
--- a/charts/llm-stack/templates/apisix-config-seeder-configmap.yaml
+++ b/charts/llm-stack/templates/apisix-config-seeder-configmap.yaml
@@ -18,12 +18,13 @@ data:
     import sys
     import json
     import os
+    from datetime import datetime
 
     url = "http://{{ include "common.names.fullname" . }}-apisix-admin-headless:9180/apisix/admin/configs"
     headers = {
         "X-API-KEY": os.getenv("ADMIN_API_KEY", ""),
         "Content-Type": "application/json",
-        "X-DIGEST": "initial-config"
+        "X-DIGEST": str(datetime.now())
     }
 
     # Wait for APISIX Admin API to be ready
@@ -62,7 +63,8 @@ data:
                 "ai-rate-limiting": {
                     "limit": {{ .Values.apisixInitialConfig.consumerUsageLimits.numTokens}},
                     "time_window": {{ .Values.apisixInitialConfig.consumerUsageLimits.timeWindowSeconds }},
-                    "limit_strategy": {{ .Values.apisixInitialConfig.consumerUsageLimits.limitStrategy | quote }}
+                    "limit_strategy": {{ .Values.apisixInitialConfig.consumerUsageLimits.limitStrategy | quote }},
+                    "rejected_code": 429
                 },
             }
         } for consumer in consumer_data
diff --git a/charts/llm-stack/values.yaml b/charts/llm-stack/values.yaml
index 2ebb3d46..70588a61 100644
--- a/charts/llm-stack/values.yaml
+++ b/charts/llm-stack/values.yaml
@@ -11,8 +11,16 @@ apisixInitialConfig:
         provider: scaleway
       - name: gemma-4-26b-a4b-it
         provider: scaleway
+      - name: glm-5.2
+        provider: scaleway
+      - name: deepseek-v4-flash-0731
+        provider: scaleway
+      - name: qwen3.5-397b-a17b
+        provider: scaleway
+      - name: mistral-medium-3.5-128b
+        provider: scaleway
   consumerUsageLimits:
-    numTokens: 10000000
+    numTokens: 50000000
     timeWindowSeconds: 86400  # 24h
     limitStrategy: total_tokens
   secrets: