Workaround for llm-uservice model ID inconsistency

Otherwise "llm-uservice" throws an exception due to None variable value, or vLLM returns error due to unrecognized model ID. Signed-off-by: Eero Tamminen <[email protected]>
opea-project · Dec 30, 2024 · 8d5a76a · 8d5a76a
1 parent 590991b
commit 8d5a76a
Showing 1 changed file with 3 additions and 0 deletions.
diff --git a/helm-charts/common/llm-uservice/templates/configmap.yaml b/helm-charts/common/llm-uservice/templates/configmap.yaml
@@ -19,7 +19,10 @@ data:
   vLLM_ENDPOINT: "http://{{ .Release.Name }}-vllm"
   {{- end }}
   {{- if .Values.LLM_MODEL_ID }}
+  # llm-uservice text gen (ChatQnA) variant
   LLM_MODEL: {{ .Values.LLM_MODEL_ID | quote}}
+  # llm-uservice DocSum & FaqGen variants
+  LLM_MODEL_ID: {{ .Values.LLM_MODEL_ID | quote}}
   {{- end }}
   HUGGINGFACEHUB_API_TOKEN: {{ .Values.global.HUGGINGFACEHUB_API_TOKEN | quote}}
   HF_HOME: "/tmp/.cache/huggingface"