🗨️Chat API

Integrate Alpha Chat in your product.

The API endpoint for Alpha Chat is compatible with OpenAI's Chat Completion API and can be used as a drop-in replacement by changing three lines of code:

Python

from openai import OpenAI

client = OpenAI(
    api_key="YOUR_TOKEN_HERE",  # Replace OpenAI's token with AlphaKEK's
    base_url="https://api.alphakek.ai/v1"  # Switch to Alpha API server
)

response = client.chat.completions.create(
  model="nexus",  # Switch to Nexus model with crypto knowledge
  messages=[
    {"role": "user", "content": "What to expect from Ethereum ETFs?"}
  ]
)

JavaScript

import OpenAI from 'openai';

const client = new OpenAI({
    apiKey: "YOUR_TOKEN_HERE", // Replace OpenAI's token with AlphaKEK's
    baseURL: 'https://api.alphakek.ai/v1' // Switch to Alpha API server
});

const response = await client.chat.completions.create({
    model: "nexus", // Switch to Nexus model with crypto knowledge
    messages: [{
        "role": "user",
        "content": "What to expect from Ethereum ETFs?"
    }]
});

Available models: versa, nexus, eclipse. Refer to AI Models for the detailed comparison of each model. The pricing information is available at #api-costs.

Chat Completions

POST/v1/chat/completions

Authorization

HTTPBearer

Body

application/json

model*ModelName

versanexuseclipse

messages*Messages

streamStream

personaPersona

stringnull

Other propertiesany

Response

200422

Successful Response

Body

application/json

id*Id

choices*Choices

created*Created

model*Model

object*Object

system_fingerprintSystem Fingerprint

stringnull

usageany of

CompletionUsagenull

Other propertiesany

Request

JavaScriptCurl

const response = await fetch('/v1/chat/completions', {
    method: 'POST',
    headers: {
      "Authorization": "Bearer <token>",
      "Content-Type": "application/json"
    },
    body: JSON.stringify({
      "model": "versa",
      "messages": [
        {
          "role": "system",
          "content": "text"
        }
      ]
    }),
});
const data = await response.json();

Response

{
  "id": "text",
  "choices": [
    {
      "finish_reason": "stop",
      "logprobs": {
        "content": [
          {
            "token": "text",
            "bytes": [],
            "logprob": 0,
            "top_logprobs": [
              {
                "token": "text",
                "bytes": [],
                "logprob": 0
              }
            ]
          }
        ]
      },
      "message": {
        "content": "text",
        "function_call": {
          "arguments": "text",
          "name": "text"
        },
        "tool_calls": [
          {
            "id": "text",
            "function": {
              "arguments": "text",
              "name": "text"
            }
          }
        ]
      }
    }
  ],
  "model": "text",
  "system_fingerprint": "text",
  "usage": {}
}

Limitations

---

• As of now, the Alpha Chat API will ignore every field not mentioned in the schema above, such as temperature or top_p. Instead, it will use the internal optimal values.

• Function calling is not supported (except for the functions of Alpha Toolkit that are being called automatically by nexus and eclipse models.

This list of limitations is subject to change.

Chat API Costs

---

The pricing for Chat API is per request and depends on the model used. Learn more about API Credits at API Pricing.

Model	API Credits per request
Versa	1
Nexus	2
Eclipse	3

📖pageAPI Usage Examples