This function processes chat completion requests by determining whether to use streaming
or non-streaming response handling based on the request payload. For streaming requests,
it configures additional options to track token usage.

## Returns

Returns a Response containing either:
- A streaming SSE connection for real-time completions
- A single JSON response for non-streaming completions

## Errors

Returns an error status code if:
- The request processing fails
- The streaming/non-streaming handlers encounter errors
- The underlying inference service returns an error

Create chat completions

Represents the chat completion choice.

This is used to represent the chat completion choice in the chat completion request.
It can be either a chat completion message or a chat completion chunk.

The reason the chat completion was finished.

The index of this choice in the list of choices.

Log probability information for the choice, if applicable.

Represents the chat completion log prob.

This is used to represent the chat completion log prob in the chat completion request.
It can be either a chat completion log prob or a chat completion log prob choice.

A list of integers representing the UTF-8 bytes representation of the token.
Useful in instances where characters are represented by multiple tokens and their byte
representations must be combined to generate the correct text representation.
Can be null if there is no bytes representation for the token.

Represents the chat completion log probs.

This is used to represent the chat completion log probs in the chat completion request.
It can be either a chat completion log probs or a chat completion log probs choice.

Represents the chat completion log probs content.

This is used to represent the chat completion log probs content in the chat completion request.
It can be either a chat completion log probs content or a chat completion log probs content choice.

A message that is part of a conversation which is based on the role
of the author of the message.

This is used to represent the message in the chat completion request.
It can be either a system message, a user message, an assistant message, or a tool message.

The role of the messages author, in this case system.

An optional name for the participant. Provides the model information to differentiate between participants of the same role.

System

The role of the messages author, in this case user.

User

The role of the messages author, in this case assistant.

The tool calls generated by the model, such as function calls.

Assistant

The role of the messages author, in this case tool.

Tool call that this message is responding to.

Tool

A named function that can be used in a chat completion.

This is used to represent the named function in the chat completion request.

A named tool choice that can be used in a chat completion.

This is used to represent the named tool choice in the chat completion request.

A function that can be used in a chat completion.

Whether to strictly validate the parameters of the function.

A tool that can be used in a chat completion.

Represents the completion usage.

This is used to represent the completion usage in the chat completion request.
It can be either a completion usage or a completion chunk usage.

Total number of tokens used (prompt + completion).

The format to return the response in.

This is used to represent the format to return the response in in the chat completion request.
It can be either text, json_object, or json_schema.

Whether to strictly validate the JSON schema.

Represents the content of a message.

This is used to represent the content of a message in the chat completion request.
It can be either a text or an array of content parts.

An array of content parts with a defined type, each can be of type text or image_url when passing in images.
You can pass multiple images by adding multiple image_url content parts. Image input is only supported when using the gpt-4o model.

Represents a part of a message content.

This is used to represent the content of a message in the chat completion request.
It can be either a text or an image.

Represents the image URL of a message content part.

This is used to represent the image URL of a message content part in the chat completion request.
It can be either a URL or a base64 encoded image data.

Either a URL of the image or the base64 encoded image data.

Specifies the stream options for the request.

If set, an additional chunk will be streamed before the data: [DONE] message.
The usage field on this chunk shows the token usage statistics for the entire request, and the choices field
will always be an empty array. All other chunks will also include a usage field, but with a null value.

Represents the tool call that the model made.

This is used to represent the tool call that the model made in the chat completion request.
It can be either a function or a tool.

The type of the tool. Currently, only function is supported.

Represents the function that the model called.

This is used to represent the function that the model called in the chat completion request.
It can be either a function or a tool call.

The arguments to call the function with, as generated by the model in JSON format.
Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema.
Validate the arguments in your code before calling your function.

A tool choice that can be used in a chat completion.

This is used to represent the tool choice in the chat completion request.
It can be either a literal tool choice or a named tool choice.

A literal tool choice that can be used in a chat completion.

This is used to represent the literal tool choice in the chat completion request.
It can be either none or auto.

Represents the create chat completion request.

This is used to represent the create chat completion request in the chat completion request.
It can be either a chat completion or a chat completion stream.
Represents the chat completion request.

This is used to represent the chat completion request in the chat completion request.
It can be either a chat completion or a chat completion stream.

bearerAuth

Represents the chat completion response.

This is used to represent the chat completion response in the chat completion request.
It can be either a chat completion or a chat completion stream.

Atoma

Welcome to Atoma - The Decentralized Private AI Network

Describes Atoma's compute AI cloud for private and verifiable AI

Compute Layer

Trust and Privacy

A brief overview of Atoma's architecture for a scalable and secure decentralized AI network

Atoma Architecture Overview

Overview

Quickstart

Learn how to use Atoma's Cloud API for AI inference

Get Started

This function processes completion requests by using the chat completions endpoint.

## Returns

Returns a Response containing either:
- A streaming SSE connection for real-time completions
- A single JSON response for non-streaming completions

## Errors

Returns an error status code if:
- The request processing fails
- The streaming/non-streaming handlers encounter errors
- The underlying inference service returns an error

Create completions

This handler processes completions requests in a confidential manner, providing additional
encryption and security measures for sensitive data processing. It supports both streaming and
non-streaming responses while maintaining data confidentiality through AEAD encryption and TEE hardware,
for full private AI compute.

## Returns

Returns a `Result` containing either:
* An HTTP response with the completions result
* A streaming SSE connection for real-time completions
* An `AtomaProxyError` error if the request processing fails

## Errors

Returns `AtomaProxyError::InvalidBody` if:
* The 'stream' field is missing or invalid in the payload

Returns `AtomaProxyError::InternalError` if:
* The inference service request fails
* Response processing encounters errors
* State manager updates fail

## Security Features

* Utilizes AEAD encryption for request/response data
* Supports TEE (Trusted Execution Environment) processing
* Implements secure key exchange using X25519
* Maintains confidentiality throughout the request lifecycle

Create confidential completions

This handler processes chat completion requests in a confidential manner, providing additional
encryption and security measures for sensitive data processing. It supports both streaming and
non-streaming responses while maintaining data confidentiality through AEAD encryption and TEE hardware,
for full private AI compute.

## Returns

Returns a `Result` containing either:
* An HTTP response with the chat completion result
* A streaming SSE connection for real-time completions
* An `AtomaProxyError` error if the request processing fails

## Errors

Returns `AtomaProxyError::InvalidBody` if:
* The 'stream' field is missing or invalid in the payload

Returns `AtomaProxyError::InternalError` if:
* The inference service request fails
* Response processing encounters errors
* State manager updates fail

## Security Features

* Utilizes AEAD encryption for request/response data
* Supports TEE (Trusted Execution Environment) processing
* Implements secure key exchange using X25519
* Maintains confidentiality throughout the request lifecycle

Create confidential chat completions

This endpoint follows the OpenAI API format for generating vector embeddings from input text,
but with confidential processing (through AEAD encryption and TEE hardware).
The handler receives pre-processed metadata from middleware and forwards the request to
the selected node.

## Returns
* `Ok(Response)` - The embeddings response from the processing node
* `Err(AtomaProxyError)` - An error status code if any step fails

## Errors
* `INTERNAL_SERVER_ERROR` - Processing or node communication failures

Create confidential embeddings

This handler processes image generation requests in a confidential manner, providing additional
encryption and security measures for sensitive data processing. It supports both streaming and
non-streaming responses while maintaining data confidentiality through AEAD encryption and TEE hardware,
for full private AI compute.

Create confidential image

This endpoint follows the OpenAI API format for generating vector embeddings from input text.
The handler receives pre-processed metadata from middleware and forwards the request to
the selected node.

# Returns
* `Ok(Response)` - The embeddings response from the processing node
* `Err(AtomaProxyError)` - An error status code if any step fails

## Errors
* `INTERNAL_SERVER_ERROR` - Processing or node communication failures

Create embeddings

Health

This endpoint processes requests to generate images using AI models by forwarding them
to the appropriate AI node. The request metadata and compute units have already been
validated by middleware before reaching this handler.

## Errors
* Returns various status codes based on the underlying `handle_image_generation_response`:
  - `INTERNAL_SERVER_ERROR` - If there's an error communicating with the AI node

Create image

This endpoint mimics the OpenAI models endpoint format, returning a list of
available models with their associated metadata. Each model includes standard
OpenAI-compatible fields to ensure compatibility with existing OpenAI client libraries.

List models

This endpoint returns a list of available models from the OpenRouter
models file. The file is expected to be in JSON format and contains
information about the models, including their IDs and other metadata.

OpenRouter models listing endpoint

This endpoint allows nodes to register or update their public address in the system.
When a node comes online or changes its address, it can use this endpoint to ensure
the system has its current address for routing requests.

## Errors

Returns various `AtomaProxyError` variants:
* `MissingHeader` - If the signature header is missing
* `InvalidHeader` - If the signature header is malformed
* `InvalidBody` - If:
  - The request body cannot be read
  - The signature is invalid
  - The body cannot be parsed
  - The sui address doesn't match the signature
* `InternalError` - If:
  - The state manager channel is closed
  - The registration event cannot be sent
  - Node Sui address lookup fails

Create node

This endpoint attempts to find a suitable node and retrieve its public key for encryption
through a two-step process:

1. First, it tries to select an existing node with a public key directly.
2. If no node is immediately available, it falls back to finding the cheapest compatible node
   and acquiring a new stack entry for it.

This endpoint is specifically designed for confidential compute scenarios where
requests need to be encrypted before being processed by nodes.

## Errors
  - `INTERNAL_SERVER_ERROR` - Communication errors or missing node public keys
  - `SERVICE_UNAVAILABLE` - No nodes available for confidential compute

Get Started

Completions

Confidential Completions

Chat

Confidential Chat

Confidential Embeddings

Confidential Images

Embeddings

Health

Images

Models

Nodes

Create chat completions

Returns

Errors

Authorizations

Body

Response