# Rasa assistant veeery slow

**URL:** <https://forum.rasa.com/t/rasa-assistant-veeery-slow/63989>\
**Category:** Rasa Open Source\
**Created:** [October 17, 2024, 8:23am UTC](https://forum.rasa.com/t/rasa-assistant-veeery-slow/63989 "2024-10-17T08:23:11Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![Gavin](https://avatars.discourse-cdn.com/v4/letter/g/439d5e/32.png) [@Gavin](https://forum.rasa.com/u/Gavin)\
**Post date:** [October 17, 2024, 8:23am UTC](https://forum.rasa.com/t/rasa-assistant-veeery-slow/63989/1 "2024-10-17T08:23:11Z")

</div>

I’m new to Rasa and have just deployed my first assistant. I’m trying to use it to apply for a job that is Rasa-based.

It works sometimes (rarely), but even when it does it’s painfully slow (up to 30 secs for a response).

It’s Rasa Pro deployed on GCP and I’m using a macbook.

Any idea why this might be and what I could do?

---

<div class="post-metadata">

**Author:** ![Gavin](https://avatars.discourse-cdn.com/v4/letter/g/439d5e/32.png) [@Gavin](https://forum.rasa.com/u/Gavin)\
**Post date:** [October 17, 2024, 4:33pm UTC](https://forum.rasa.com/t/rasa-assistant-veeery-slow/63989/2 "2024-10-17T16:33:09Z")

</div>

I’ve now removed all custom actions and run debug. This is the debug log:

/Users/gavinreid/Code/rasa\_projects/cv\_bot/p39/lib/python3.9/site-packages/rasa/core/tracker\_store.py:1044: MovedIn20Warning: Deprecated API features detected! These feature(s) are not compatible with SQLAlchemy 2.0. To prevent incompatible upgrades prior to updating applications, ensure requirements files are pinned to “sqlalchemy\<2.0”. Set environment variable SQLALCHEMY\_WARN\_20=1 to show all deprecation warnings. Set environment variable SQLALCHEMY\_SILENCE\_UBER\_WARNING=1 to silence this message. (Background on SQLAlchemy 2.0 at: [Error Messages — SQLAlchemy 2.0 Documentation](https://sqlalche.me/e/b8d9)) Base: DeclarativeMeta = declarative\_base() /Users/gavinreid/Code/rasa\_projects/cv\_bot/p39/lib/python3.9/site-packages/rasa/shared/utils/validation.py:134: DeprecationWarning: pkg\_resources is deprecated as an API. See [Package Discovery and Resource Access using pkg\_resources - setuptools 75.2.0.post20241017 documentation](https://setuptools.pypa.io/en/latest/pkg_resources.html) import pkg\_resources /Users/gavinreid/Code/rasa\_projects/cv\_bot/p39/lib/python3.9/site-packages/pkg\_resources/ **init**.py:3154: DeprecationWarning: Deprecated call to pkg\_resources.declare\_namespace(‘mpl\_toolkits’). Implementing implicit namespace packages (as specified in PEP 420) is preferred to pkg\_resources.declare\_namespace. See [Keywords - setuptools 75.2.0.post20241017 documentation](https://setuptools.pypa.io/en/latest/references/keywords.html#keyword-namespace-packages) declare\_namespace(pkg) /Users/gavinreid/Code/rasa\_projects/cv\_bot/p39/lib/python3.9/site-packages/pkg\_resources/ **init**.py:3154: DeprecationWarning: Deprecated call to pkg\_resources.declare\_namespace(‘ruamel’). Implementing implicit namespace packages (as specified in PEP 420) is preferred to pkg\_resources.declare\_namespace. See [Keywords - setuptools 75.2.0.post20241017 documentation](https://setuptools.pypa.io/en/latest/references/keywords.html#keyword-namespace-packages) declare\_namespace(pkg) 2024-10-17 18:28:56 DEBUG rasa.cli.utils - Parameter ‘credentials’ not set. Using default location ‘credentials.yml’ instead. /Users/gavinreid/Code/rasa\_projects/cv\_bot/p39/lib/python3.9/site-packages/sanic\_cors/extension.py:39: DeprecationWarning: distutils Version classes are deprecated. Use packaging.version instead. SANIC\_VERSION = LooseVersion(sanic\_version) 2024-10-17 18:28:58 DEBUG h5py.\_conv - Creating converter from 7 to 5 2024-10-17 18:28:58 DEBUG h5py.\_conv - Creating converter from 5 to 7 2024-10-17 18:28:58 DEBUG h5py.\_conv - Creating converter from 7 to 5 2024-10-17 18:28:58 DEBUG h5py.\_conv - Creating converter from 5 to 7 2024-10-17 18:28:58 DEBUG jax.\_src.path - etils.epath was not found. Using pathlib for file I/O. /Users/gavinreid/Code/rasa\_projects/cv\_bot/p39/lib/python3.9/site-packages/tensorflow/lite/python/util.py:52: DeprecationWarning: jax.xla\_computation is deprecated. Please use the AOT APIs. from jax import xla\_computation as \_xla\_computation 2024-10-17 18:28:59 DEBUG sanic.root - Sanic-CORS: Configuring CORS with resources: {‘/_‘: {‘origins’: [’‘], ‘methods’: ‘DELETE, GET, HEAD, OPTIONS, PATCH, POST, PUT’, ‘allow\_headers’: [’._’], ‘expose\_headers’: ‘filename’, ‘supports\_credentials’: True, ‘max\_age’: None, ‘send\_wildcard’: False, ‘automatic\_options’: True, ‘vary\_header’: True, ‘resources’: {‘/\*’: {‘origins’: ‘’}}, ‘intercept\_exceptions’: True, ‘always\_send’: True}} 2024-10-17 18:28:59 DEBUG rasa.core.utils - Available web server routes: /webhooks/rasa GET rasa\_core\_no\_api.custom\_webhook\_RasaChatInput.health /webhooks/rasa/webhook POST rasa\_core\_no\_api.custom\_webhook\_RasaChatInput.receive /webhooks/rest GET rasa\_core\_no\_api.custom\_webhook\_RestInput.health /webhooks/rest/webhook POST rasa\_core\_no\_api.custom\_webhook\_RestInput.receive /socket.io GET rasa\_core\_no\_api.handle\_request / GET rasa\_core\_no\_api.hello /webhooks/socketio GET rasa\_core\_no\_api.socketio\_webhook.health 2024-10-17 18:28:59 INFO root - Starting Rasa server on [http://0.0.0.0:5005](http://0.0.0.0:5005) 2024-10-17 18:28:59 DEBUG rasa.core.utils - Using the default number of Sanic workers (1). 2024-10-17 18:28:59 DEBUG rasa.telemetry - Skipping telemetry reporting: no license hash found. 2024-10-17 18:28:59 DEBUG rasa.core.tracker\_store - Connected to InMemoryTrackerStore. 2024-10-17 18:28:59 DEBUG rasa.core.lock\_store - Connected to lock store ‘InMemoryLockStore’. 2024-10-17 18:28:59 DEBUG rasa.core.nlg.generator - Instantiated NLG to ‘TemplatedNaturalLanguageGenerator’. 2024-10-17 18:28:59 INFO rasa.core.processor - Loading model models/20241017-182848-espressivo-ladder.tar.gz… 2024-10-17 18:28:59 DEBUG rasa.engine.storage.local\_model\_storage - Extracted model to ‘/var/folders/gw/1k7\_nn090cd\_fttzv0f5lylm0000gn/T/tmp39drr5cw’. 2024-10-17 18:28:59 DEBUG rasa.engine.graph - Node ‘nlu\_message\_converter’ loading ‘NLUMessageConverter.load’ and kwargs: ‘{}’. 2024-10-17 18:28:59 DEBUG rasa.engine.graph - Node ‘run\_WhitespaceTokenizer0’ loading ‘WhitespaceTokenizer.load’ and kwargs: ‘{}’. 2024-10-17 18:28:59 DEBUG rasa.engine.graph - Node ‘run\_CountVectorsFeaturizer1’ loading ‘CountVectorsFeaturizer.load’ and kwargs: ‘{}’. 2024-10-17 18:28:59 DEBUG rasa.engine.storage.local\_model\_storage - Resource ‘train\_CountVectorsFeaturizer1’ was requested for reading. 2024-10-17 18:28:59 DEBUG rasa.engine.graph - Node ‘run\_DIETClassifier2’ loading ‘DIETClassifier.load’ and kwargs: ‘{}’. 2024-10-17 18:28:59 DEBUG rasa.engine.storage.local\_model\_storage - Resource ‘train\_DIETClassifier2’ was requested for reading. 2024-10-17 18:28:59 DEBUG rasa.utils.tensorflow.models - Loading the model from /var/folders/gw/1k7\_nn090cd\_fttzv0f5lylm0000gn/T/tmp3ryhzwyd/train\_DIETClassifier2/DIETClassifier.tf\_model with finetune\_mode=False… 2024-10-17 18:28:59 DEBUG rasa.nlu.classifiers.diet\_classifier - You specified ‘DIET’ to train entities, but no entities are present in the training data. Skipping training of entities. 2024-10-17 18:28:59 DEBUG rasa.nlu.classifiers.diet\_classifier - Following metrics will be logged during training: 2024-10-17 18:28:59 DEBUG rasa.nlu.classifiers.diet\_classifier - t\_loss (total loss) 2024-10-17 18:28:59 DEBUG rasa.nlu.classifiers.diet\_classifier - i\_acc (intent acc) 2024-10-17 18:28:59 DEBUG rasa.nlu.classifiers.diet\_classifier - i\_loss (intent loss) 2024-10-17 18:28:59 WARNING absl - At this time, the v2.11+ optimizer tf.keras.optimizers.Adam runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at tf.keras.optimizers.legacy.Adam. 2024-10-17 18:28:59 WARNING absl - There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., tf.keras.optimizers.legacy.Adam. 2024-10-17 18:29:03 DEBUG rasa.utils.tensorflow.models - Finished loading the model. /Users/gavinreid/Code/rasa\_projects/cv\_bot/p39/lib/python3.9/site-packages/rasa/utils/train\_utils.py:530: UserWarning: constrain\_similarities is set to False. It is recommended to set it to True when using cross-entropy loss. rasa.shared.utils.io.raise\_warning( 2024-10-17 18:29:03 DEBUG rasa.engine.graph - Node ‘run\_RegexMessageHandler’ loading ‘RegexMessageHandler.load’ and kwargs: ‘{}’. 2024-10-17 18:29:03 DEBUG rasa.engine.graph - Node ‘domain\_provider’ loading ‘DomainProvider.load’ and kwargs: ‘{}’. 2024-10-17 18:29:03 DEBUG rasa.engine.storage.local\_model\_storage - Resource ‘domain\_provider’ was requested for reading. 2024-10-17 18:29:03 DEBUG rasa.engine.graph - Node ‘run\_MemoizationPolicy0’ loading ‘MemoizationPolicy.load’ and kwargs: ‘{}’. 2024-10-17 18:29:03 DEBUG rasa.engine.storage.local\_model\_storage - Resource ‘train\_MemoizationPolicy0’ was requested for reading. 2024-10-17 18:29:03 DEBUG rasa.engine.graph - Node ‘run\_RulePolicy1’ loading ‘RulePolicy.load’ and kwargs: ‘{}’. 2024-10-17 18:29:03 DEBUG rasa.engine.storage.local\_model\_storage - Resource ‘train\_RulePolicy1’ was requested for reading. 2024-10-17 18:29:03 DEBUG rasa.engine.graph - Node ‘run\_TEDPolicy2’ loading ‘TEDPolicy.load’ and kwargs: ‘{}’. 2024-10-17 18:29:03 DEBUG rasa.engine.storage.local\_model\_storage - Resource ‘train\_TEDPolicy2’ was requested for reading. 2024-10-17 18:29:03 DEBUG rasa.utils.tensorflow.models - Loading the model from /var/folders/gw/1k7\_nn090cd\_fttzv0f5lylm0000gn/T/tmp3ryhzwyd/train\_TEDPolicy2/ted\_policy.tf\_model with finetune\_mode=False… 2024-10-17 18:29:03 WARNING absl - At this time, the v2.11+ optimizer tf.keras.optimizers.Adam runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at tf.keras.optimizers.legacy.Adam. 2024-10-17 18:29:03 WARNING absl - There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., tf.keras.optimizers.legacy.Adam. 2024-10-17 18:29:06 DEBUG rasa.utils.tensorflow.models - Finished loading the model. /Users/gavinreid/Code/rasa\_projects/cv\_bot/p39/lib/python3.9/site-packages/rasa/shared/utils/io.py:99: UserWarning: ‘evaluate\_every\_number\_of\_epochs=20’ is greater than ‘epochs=1’. No evaluation will occur. 2024-10-17 18:29:06 DEBUG rasa.engine.graph - Node ‘rule\_only\_data\_provider’ loading ‘RuleOnlyDataProvider.load’ and kwargs: ‘{}’. 2024-10-17 18:29:06 DEBUG rasa.engine.storage.local\_model\_storage - Resource ‘train\_RulePolicy1’ was requested for reading. 2024-10-17 18:29:06 DEBUG rasa.engine.graph - Node ‘select\_prediction’ loading ‘DefaultPolicyPredictionEnsemble.load’ and kwargs: ‘{}’. 2024-10-17 18:29:06 INFO root - Rasa server is up and running. 2024-10-17 18:29:06 INFO root - Enabling coroutine debugging. Loop id 6126255648.

---

<div class="post-metadata">

**Author:** ![ttkrpink](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.rasa.com/ttkrpink/32/19523_2.png) [@ttkrpink](https://forum.rasa.com/u/ttkrpink)\
**Post date:** [July 13, 2025, 4:37pm UTC](https://forum.rasa.com/t/rasa-assistant-veeery-slow/63989/3 "2025-07-13T16:37:38Z")

</div>

Same problem. Extremely slow. Local machine python installation.

Codespace speed is fine.

Did you solve it?