How AI is applied across API Evangelist and APIs.io. Read my AI disclosure →
API Evangelist API Evangelist
Discovery
Learnings
Guidance
Toolbox
Alignment
API Evangelist LLC

Letta Conversations API

Conversational session threads anchored to an agent.

Letta Conversations API is one of 41 APIs that Letta publishes on the APIs.io network, described by a machine-readable OpenAPI specification.

This API exposes 14 JSON Schema definitions.

Tagged areas include Conversations. The published artifact set on APIs.io includes an OpenAPI specification, API documentation, pricing, a changelog, and 14 JSON Schemas.

This API exposes 13 operations across 9 paths, and defines 82 schemas. It is described by OpenAPI 3.2.0, at version 1.0.0.

Requests are made against 3 base URLs: https://api.letta.com, https://app.letta.com, http://localhost:8283.

13 operations 9 paths 82 schemas 1 DELETE3 GET1 PATCH8 POST

Metadata

The identity and technical contract details declared by the specification.

Specification
OpenAPI 3.2.0
API Version
1.0.0
Base URL
https://api.letta.com
Authentication
HTTP Bearer
Contact
License
Resource Areas
1

Authentication & Security 1

Letta Conversations API declares 1 security scheme for authenticating requests. It accepts HTTP bearer tokens (bearerAuth). By default, every request must be authenticated.

Paths & Operations 13

Across 9 paths, the API surfaces 13 operations — 1 DELETE, 3 GET, 1 PATCH, 8 POST. Each is listed below with its method, path, parameters, and response codes.

Conversations 13

Conversational session threads anchored to an agent.

POST
/v1/conversations/
Create Conversation
create_conversation 1 param body → 200422
GET
/v1/conversations/
List Conversations
list_conversations 6 params → 200422
GET
/v1/conversations/{conversation_id}
Retrieve Conversation
retrieve_conversation 1 param → 200422
PATCH
/v1/conversations/{conversation_id}
Update Conversation
update_conversation 1 param body → 200422
DELETE
/v1/conversations/{conversation_id}
Delete Conversation
delete_conversation 1 param → 200422
POST
/v1/conversations/{conversation_id}/fork
Fork Conversation
fork_conversation 2 params → 200422
GET
/v1/conversations/{conversation_id}/messages
List Conversation Messages
list_conversation_messages 10 params → 200422
POST
/v1/conversations/{conversation_id}/messages
Send Conversation Message
send_conversation_message 1 param body → 200422
POST
/v1/conversations/{conversation_id}/messages/preview-raw-payload
Preview Conversation Model Request
preview_conversation_model_request 1 param body → 200422
POST
/v1/conversations/{conversation_id}/stream
Retrieve Conversation Stream
retrieve_conversation_stream 1 param body → 200422
POST
/v1/conversations/{conversation_id}/cancel
Cancel Conversation
cancel_conversation 2 params → 200422
POST
/v1/conversations/{conversation_id}/recompile
Recompile Conversation
recompile_conversation 2 params body → 200422
POST
/v1/conversations/{conversation_id}/compact
Compact Conversation
compact_conversation 1 param body → 200422

Schemas 82

The contract defines 82 schemas that model the data the API accepts and returns. The most detailed are ConversationMessageRequest (21 properties), ToolReturnMessage (16 properties), ApprovalResponseMessage (14 properties), ReasoningMessage (13 properties). Each schema is shown below with its type and property counts.

ToolReturnContent
object
4 properties 3 required
ZAIModelSettings
object
Z.ai (ZhipuAI) model configuration (OpenAI-compatible).
6 properties
OpenAIReasoning
object
1 property
GoogleAIModelSettings
object
6 properties
OpenRouterModelSettings
object
OpenRouter model configuration (OpenAI-compatible).
5 properties
ImageContent
object
2 properties 1 required
ToolCall
object
3 properties 3 required
ClientToolSchema
object
Schema for a client-side tool passed in the request. Client-side tools are executed by the client, not the server. When the agent calls a client-side tool, exe…
3 properties 1 required
BasetenModelSettings
object
Baseten model configuration (OpenAI-compatible).
4 properties
ConversationMessageRequest
object
Request for sending messages to a conversation. Streams by default.
21 properties
XAIModelSettings
object
xAI model configuration (OpenAI-compatible).
5 properties
RedactedReasoningContent
object
Sent via the Anthropic Messages API
2 properties 1 required
LettaMessageContentUnion
MessageCreate
object
Request to create a message
8 properties 2 required
RetrieveStreamRequest
object
7 properties
GeminiThinkingConfig
object
2 properties
Conversation
object
Represents a conversation on an agent for concurrent messaging.
12 properties 2 required
ClientSkillSchema
object
Schema for a client-side skill passed in the request. Client-side skills represent environment-provided capabilities (e.g. project-scoped skills) that are not…
3 properties 3 required
JsonObjectResponseFormat
object
Response format for JSON object responses.
1 property
ReasoningMessage
object
Representation of an agent's internal reasoning. Args: id (str): The ID of the message date (datetime): The date the message was created in ISO format name (Op…
13 properties 3 required
ToolReturnCreate
object
Submit tool return(s) from client-side tool execution. This is the preferred way to send tool results back to the agent after client-side tool execution. It is…
4 properties 1 required
AnthropicModelSettings
object
9 properties
ToolReturnMessage
object
A message representing the return value of a tool call (generated by Letta executing the requested tool). Args: id (str): The ID of the message date (datetime)…
16 properties 5 required
SummarizedReasoningContent
object
The style of reasoning content returned by the OpenAI Responses API
4 properties 2 required
letta__schemas__openai__chat_completion_response__ChatCompletionTokenLogprob
object
4 properties 3 required
LettaUserMessageContentUnion
BedrockModelSettings
object
AWS Bedrock model configuration.
5 properties
CompactionStats
object
Statistics about a memory compaction operation.
6 properties 4 required
ReasoningContent
object
Sent via the Anthropic Messages API
4 properties 2 required
UpdateConversation
object
Request model for updating a conversation.
4 properties
ToolCallContent
object
5 properties 3 required
LettaStopReason
object
The stop reason from Letta indicating why agent loop stopped execution.
2 properties 1 required
GroqModelSettings
object
Groq model configuration (OpenAI-compatible).
5 properties
CompactionResponse
object
3 properties 3 required
ApprovalCreate
object
Input to approve or deny a tool call request
7 properties
letta__schemas__openai__chat_completion_response__ChoiceLogprobs
object
2 properties
LettaToolReturnContentUnion
OpenAIModelSettings
object
7 properties
AzureModelSettings
object
Azure OpenAI model configuration (OpenAI-compatible).
5 properties
TurnTokenData
object
Token data for a single LLM generation turn in a multi-turn agent interaction. Used for RL training to track token IDs and logprobs across all LLM calls, not j…
5 properties 1 required
LettaErrorMessage
object
Error messages are used to notify the client of an error that occurred during the agent's execution.
6 properties 4 required
UserMessage
object
A message sent by the user. Never streamed back on a response, only used for cursor pagination. Args: id (str): The ID of the message date (datetime): The date…
11 properties 3 required
ValidationError
object
3 properties 3 required
SystemMessage
object
A message generated by the system. Never streamed back on a response, only used for cursor pagination. Args: id (str): The ID of the message date (datetime): T…
11 properties 3 required
UrlImage
object
2 properties 1 required
ApprovalRequestMessage
object
A message representing a request for approval to call a tool (generated by the LLM to trigger tool execution). Args: id (str): The ID of the message date (date…
12 properties 3 required
Base64Image
object
4 properties 2 required
letta__schemas__letta_message__ToolReturn
object
6 properties 3 required
EventMessage
object
A message for notifying the developer that an event that has occured (e.g. a compaction). Events are NOT part of the context window.
12 properties 4 required
letta__server__rest_api__routers__v1__conversations__CompactionRequest
object
2 properties
HiddenReasoningMessage
object
Representation of an agent's internal reasoning where reasoning content has been hidden from the response. Args: id (str): The ID of the message date (datetime…
12 properties 3 required
TextResponseFormat
object
Response format for plain text responses.
1 property
TogetherModelSettings
object
Together AI model configuration (OpenAI-compatible).
5 properties
CompactionSettings-Input
object
Configuration for conversation compaction / summarization. Per-model settings (temperature, max tokens, etc.) are derived from the default configuration for th…
7 properties
MessageType
string
ToolCallMessage
object
A message representing a request to call a tool (generated by the LLM to trigger tool execution). Args: id (str): The ID of the message date (datetime): The da…
12 properties 3 required
SummaryMessage
object
A message representing a summary of the conversation. Sent to the LLM as a user or system message depending on the provider.
12 properties 3 required
LettaResponse
object
Response object from an agent interaction, consisting of the new messages generated by the agent and usage statistics. The type of the returned messages can be…
5 properties 3 required
ChatGPTOAuthReasoning
object
Reasoning configuration for ChatGPT OAuth models (GPT-5.x, o-series).
1 property
TextContent
object
3 properties 1 required
LettaMessageUnion
SummarizedReasoningContentPart
object
2 properties 2 required
GoogleVertexModelSettings
object
6 properties
HTTPValidationError
object
1 property
ApprovalReturn
object
4 properties 2 required
AssistantMessage
object
A message sent by the LLM in response to user input. Used in the LLM context. Args: id (str): The ID of the message date (datetime): The date the message was c…
11 properties 3 required
AnthropicThinking
object
2 properties
SGLangModelSettings
object
SGLang model configuration (OpenAI-compatible runtime with SGLang-specific parsing).
8 properties
LettaPing
object
A ping message used as a keepalive to prevent SSE streams from timing out during long running requests. Args: id (str): The ID of the message date (datetime):…
10 properties 2 required
JsonSchemaResponseFormat
object
Response format for JSON schema-based responses.
2 properties 1 required
ToolCallDelta
object
3 properties
ChatGPTOAuthModelSettings
object
ChatGPT OAuth model configuration (uses ChatGPT backend API).
5 properties
LettaAssistantMessageContentUnion
LettaImage
object
5 properties 1 required
DeepseekModelSettings
object
Deepseek model configuration (OpenAI-compatible).
5 properties
letta__schemas__openai__chat_completion_response__TopLogprob
object
3 properties 2 required
LettaUsageStatistics
object
Usage statistics for the agent interaction. Attributes: completiontokens (int): The number of tokens generated by the agent. prompttokens (int): The number of…
10 properties
ZAIThinking
object
Thinking configuration for ZAI GLM-4.5+ models.
2 properties
StopReasonType
string
CreateConversation
object
Request model for creating a new conversation.
4 properties
OmittedReasoningContent
object
A placeholder for reasoning content we know is present, but isn't returned by the provider (e.g. OpenAI GPT-5 on ChatCompletions)
2 properties
ApprovalResponseMessage
object
A message representing a response form the user indicating whether a tool has been approved to run. Args: id (str): The ID of the message date (datetime): The…
14 properties 2 required

Specification

The full machine-readable OpenAPI contract behind this narrative.

Source

letta-conversations-api-openapi.yml Raw ↑

Other APIs Letta publishes across the network.

Letta Code
Letta Admin API
Letta Agents API
Letta Archives API
Letta Blocks API
Letta Client-Side Access Tokens API
Letta Device Storage API
Letta Embeddings API
Letta Environments API
Letta Feeds API
Letta Folders API
Letta Groups API
Where this information came from

This is an independent, third-party profile of Letta Conversations API, published by API Evangelist. We do not operate, host, resell, or support these APIs, and we are not affiliated with or endorsed by the company unless stated above. Everything here is built from publicly available information — the company's own site, developer portal, documentation, public repositories, and the specifications it publishes for public use. Nothing is obtained by breaching a system, defeating an access control, or using credentials.

The Kin Score and Agent Readiness rating are independently calculated assessments of a company's public API artifacts, scored against a published rubric. They are not certifications, endorsements, security assessments, or audits.

Corrections, re-scores, and removal are free — no partnership or purchase required, and you do not need to justify the request. A removed company is recorded as unrated, never scored zero for having asked. Acknowledgement within one business day; removal within two.

info@apievangelist.com · Read the full data-sourcing policy →
On a security or compliance team? Put security in the subject line and you will get a person, not a form — we will tell you exactly which public URLs this profile was built from.