Quotas and limits

This document contains the commonly encountered quotas and limits for use of Dialogflow. Information on how to find a complete list of quotas and limits is provided below. We reserve the right to change these constraints, and this page will be updated to reflect any changes.

Quotas

Quotas are default constraints applied to your project. If you are using a paid edition, you can request a quota increase. The purpose of quotas is to throttle service load per client, which protects a service from being overloaded and a client from unexpected resource usage.

Dialogflow quotas vary, depending on the edition of your agent. Quotas apply to each project and are shared across all applications and IP addresses using a project.

Quota time boundary details:

  • Per-month quotas are replenished on the 1st of each month at 12:00 AM Pacific Time.
  • Per-day quotas are replenished daily at 12:00 AM Pacific Time. When a Pacific Time daylight savings transition increases the length of a day, a proportional quota bonus will be granted. However, there is no equivalent deduction when a shorter day occurs due to the daylight savings transition.
  • Per-minute quotas are refreshed every 60 seconds on the minute.

The quotas listed below are the commonly encountered quotas. To review all quotas, see the Google Cloud Console Dialogflow quotas page.

Quota increase request

You can request a quota increase if necessary.

If you use different projects for Dialogflow agents and for authenticating with service accounts, request a quota increase for the consumer project. For more information, see Using multiple projects for ES or Using multiple projects for CX.

For ES agents, make sure to upgrade the agent to the Dialogflow Essentials edition before requesting a quota increase.

Quotas table

The following terms are used to describe pricing and quotas:

  • Request: A request is defined as any API call to the Dialogflow service, whether direct with API usage or indirect with integration or console usage. Depending on the task and design of the agent, the number of requests needed for an end-user to accomplish a task with a Dialogflow agent can vary greatly.
  • Session: A session is a conversation between an end-user and a Dialogflow agent. A session remains active and its data is stored for 30 minutes after the last request is sent for the session. A session can be either a chat session or a voice session.
    • Chat session: A chat session only uses text for both requests and responses.
    • Voice session: A voice session uses audio for requests, responses, or both.
  • Consumer projects and resource projects: If you use multiple projects, it is possible that the project associated with your request authentication (consumer project) is not the same project that is associated with the agent in the request (resource project). In this case, the consumer project is used to determine prices and quotas. For more information, see Using multiple projects for ES or Using multiple projects for CX.

The following tables provide a quota comparison for editions by agent type. Unless a feature is indicated as included, quotas are cumulative for all features used by a request.

CX Agent

Feature CX Edition
Text
(includes all DetectIntent, StreamingDetectIntent, and FulfillIntent requests that do not contain audio)
1200 requests per minute
Audio input/output
(speech recognition, speech-to-text, STT, speech synthesis, text-to-speech, TTS, telephony)
600 requests per minute
Dialogflow phone gateway (Preview)
Includes audio input and output.
100 total phone minutes per minute ‡
Design-time write requests
For example, calls to build or update an agent.
60 requests per minute
Design-time read requests
For example, calls to list or get agent resources.
200 requests per minute
Other session requests
For example, setting or getting session entities.
100 requests per minute

ES Agent

Feature Trial Edition Essentials Edition
Text
(includes all DetectIntent and StreamingDetectIntent requests that do not contain audio)
180 requests per minute ¶ 600 requests per minute ¶
Audio input
(also known as speech recognition, speech-to-text, STT)
100 requests per minute †

1000 requests per day †

15,000 requests per month †

Maximum 60 seconds of audio length per request †
300 requests per minute †

Maximum 60 seconds of audio length per request †
Audio output
(also known as speech synthesis, text-to-speech, TTS)
Same as audio input Same as audio input
Knowledge connectors (Beta) Maximum 10 MB total document size

1000 requests per month

100 requests per day
Unlimited #
Sentiment analysis Not available Unlimited #
Dialogflow phone gateway (Preview)
Includes audio input and output.
Toll-free number: Not available

3 total phone minutes per minute ‡

30 phone minutes per day

500 phone minutes per month

Phone number reserved for 30 days
100 total phone minutes per minute ‡
Mega agent 1000 requests per day Unlimited #
Design-time write requests
For example, calls to build or update an agent.
60 requests per minute 60 requests per minute
Design-time read requests
For example, calls to list or get agent resources.
60 requests per minute 60 requests per minute
Other session requests
For example, setting or getting session entities or updating/querying context.
100 requests per minute 100 requests per minute

Agent Assist

If you are using Agent Assist, the quotas listed under the CX Agent and ES Agent tabs also apply, according to which agent type you're using.

Feature Trial Edition Essentials Edition
Agent Assist conversation other operations
Includes all requests that manage Conversations and Participants.
Not available 300 requests per minute
Agent Assist analyze text/audio operations
Includes all AnalyzeContent or StreamingAnalyzeContent requests for any conversation stage. This includes all such requests involving text, speech synthesis, and/or speech recognition.
Not available 300 requests per minute
Agent Assist conversation summary suggestion polling requests
Includes all requests for getting conversation summaries.
Not available 300 requests per minute

Limits

Limits are fixed constraints, which cannot be increased. Many resources and fields have count, duration, or length limitations, which are fixed constraints for the service implementation.

The following tables list commonly encountered limits. Feature-specific documentation and API reference documentation may provide additional limits.

Count limits

CX Agent

Description Limit
Maximum number of agents per Google Cloud project 1000
Maximum number of flows per agent 50
Maximum number of entity types per agent 250
Maximum number of webhooks per agent 100
Maximum number of environments per agent 20
Maximum number of test cases per agent 2000
Maximum number of changelog per agent 20,000
Maximum number of pages per flow 250
Maximum number of route groups per flow 100
Maximum number of versions per flow 20
Maximum number of intents per agent 10,000
Maximum number of referenced intents per flow 2000
Maximum number of training phrases per intent and per language (except "Default Negative Intent") 2000
Maximum number of training phrases per flow and per language 100,000
Maximum number of parameters per intent 20
Maximum number of entity entries per entity 30,000
Maximum number of entity synonyms per entity entry 200
Maximum number of entity reference values and synonyms per agent and per language 1,000,000
Maximum number of parameters per page 20
Maximum number of routes per page 2000
Maximum number of event handlers per page 100
Maximum number of routes per route group 2000

ES Agent

Description Limit
Maximum number of agents per region and per Google Cloud project ¶ 1
Maximum number of intents 2000
Maximum number of entity types 250
Maximum number of training phrases per intent and per language 2000
Maximum number of training phrases per agent and per language 100,000
Maximum number of entity entries 30000
Maximum number of entity synonyms per entity entry 200
Maximum number of entity reference values and synonyms per agent and per language 1,000,000
Maximum number of parameters per intent 20
Maximum number of input contexts per intent 5
Maximum number of output contexts per intent 30
Maximum number of text responses per intent and per language 30
Maximum number of prompts per parameter and per language 30
Maximum number of environments per agent 10
Maximum number of versions per agent 1000

Agent Assist

If you are using Agent Assist, the limits listed under the CX Agent and ES Agent tabs also apply according to which agent type you're using.

Description Limit
Minimum number of conversations required for model training 30,000
Maximum number of models trained concurrently per project 1
Maximum number of training node GPU hours per month 4200

Duration limits

CX Agent

Description Limit
Maximum timeout per webhook 30 seconds
Maximum audio input duration 120 seconds

ES Agent

Description Limit
Maximum webhook timeout 5 seconds
Maximum telephony call duration 3.5 hours

Length limits

CX Agent

Description Limit
Maximum training phrase length 768 characters
Maximum text response length 4,000 characters
Maximum text input length for detect intent matched by non-generative intents or parameters 256 characters

ES Agent

Description Limit
Minimum agent name length 1 character
Maximum agent name length 150 characters
Maximum agent description length 500 characters
Maximum intent name length 100 characters
Maximum entity type name length 30 characters
Maximum training phrase length 768 characters
Maximum action name length 250 characters
Maximum context name length 250 characters
Maximum entry value length 512 characters
Maximum parameter name length 30 characters
Maximum text response length 4000 characters
Maximum detect intent text input length * 256 characters
Maximum event name length 150 characters

Size limits

CX Agent

Description Limit
Maximum agent size for agent export/restore from/to a local file system * ~10 MB
Maximum speech synthesis request size See the TTS API content limits

ES Agent

Description Limit
Maximum agent size (unzipped content) for agent import/restore 50 MB
Maximum speech synthesis request size See the TTS API content limits