Confidential Chat

Create confidential chat completion

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

POST

confidential

chat

completions

import { AtomaSDK } from "atoma-sdk";

const atomaSDK = new AtomaSDK({
  bearerAuth: process.env["ATOMASDK_BEARER_AUTH"] ?? "",
});

async function run() {
  const completion = await atomaSDK.confidentialChat.create({
    messages: [
      {"role": "developer", "content": "You are a helpful assistant."},
      {"role": "user", "content": "Hello!"}
    ],
    model: "meta-llama/Llama-3.3-70B-Instruct"
  });

  console.log(completion.choices[0]);
}

run();

Authorizations

Authorization

string

header

required

Bearer authentication header of the form Bearer <token>, where <token> is your auth token.

Body

application/json

A request for confidential computation that includes encrypted data and associated cryptographic parameters

ciphertext

string

required

The encrypted payload that needs to be processed (base64 encoded)

client_dh_public_key

string

required

Client's public key for Diffie-Hellman key exchange (base64 encoded)

model_name

string

required

Model name

node_dh_public_key

string

required

Node's public key for Diffie-Hellman key exchange (base64 encoded)

nonce

string

required

Cryptographic nonce used for encryption (base64 encoded)

plaintext_body_hash

string

required

Hash of the original plaintext body for integrity verification (base64 encoded)

salt

string

required

Salt value used in key derivation (base64 encoded)

stack_small_id

integer

required

Unique identifier for the small stack being used

Required range: x > 0

num_compute_units

integer | null

Number of compute units to be used for the request, for image generations, as this value is known in advance (the number of pixels to generate)

Required range: x > 0

stream

boolean | null

Indicates whether this is a streaming request

Response

200 - application/json

Represents a response from a confidential compute request

ciphertext

string

required

Encrypted response body (base64 encoded)

nonce

string

required

Nonce used for encryption (base64 encoded)

response_hash

string | null

Hash of the response body (base64 encoded)

signature

string | null

Signature of the response body (base64 encoded)

usage

object | null

Usage statistics for the request

Create chat completion Create confidential embeddings

import { AtomaSDK } from "atoma-sdk";

const atomaSDK = new AtomaSDK({
  bearerAuth: process.env["ATOMASDK_BEARER_AUTH"] ?? "",
});

async function run() {
  const completion = await atomaSDK.confidentialChat.create({
    messages: [
      {"role": "developer", "content": "You are a helpful assistant."},
      {"role": "user", "content": "Hello!"}
    ],
    model: "meta-llama/Llama-3.3-70B-Instruct"
  });

  console.log(completion.choices[0]);
}

run();

Get Started

Chat

Confidential Chat

Confidential Embeddings

Confidential Images

Embeddings

Health

Images

Models

Nodes

Create confidential chat completion

Returns

Errors

Security Features

Authorizations

Body

Response