Rasa Is Consuming a lot of Memory

Hello Rasa Team,

We are using RASA 1.10 and we are in production with our client. It is observed after too many requests the instance is consuming a lot of memory in RAM. approximately 100 request under 5 minutes is resulting 1 GB of memory.

Is tensorflow consuming a portion of RAM for every request? Your input will really help us with further troubleshooting.


Configuration for Rasa NLU.


language: en pipeline:

  • name: WhitespaceTokenizer
  • name: RegexFeaturizer
  • name: LexicalSyntacticFeaturizer
  • name: CountVectorsFeaturizer
  • name: CountVectorsFeaturizer analyzer: “char_wb” min_ngram: 1 max_ngram: 4
  • name: DIETClassifier epochs: 100
  • name: EntitySynonymMapper
  • name: ResponseSelector epochs: 100

Configuration for Rasa Core.



  • name: “FallbackPolicy” nlu_threshold: 0.4 core_threshold: 0.3 fallback_action_name: “action_default_fallback”
  • name: MemoizationPolicy
  • name: TEDPolicy max_history: 5 epochs: 100
  • name: MappingPolicy
  • name: “FormPolicy”

I am Having a same type of issue. The problem is that, by default, it allocates the full amount of available memory when it is launched. I see that all 12 GB of the memory is used up.