Skip to main content
POST
/
api
/
dataframer
/
evaluations
/
{evaluation_id}
/
chat
JavaScript
import Dataframer from 'dataframer';

const client = new Dataframer({
  apiKey: 'My API Key',
});

const response = await client.dataframer.evaluations.chat.sendMessage('evaluation_id', {
  user_message: 'Why did some samples score lower than others?',
});

console.log(response.evaluation_id);
{
  "evaluation_id": "3c90c3cc-0d44-4b50-8888-8dd25736052a",
  "user_message": "<string>",
  "assistant_message": "<string>",
  "id": "3c90c3cc-0d44-4b50-8888-8dd25736052a",
  "user": 123,
  "user_email": "<string>",
  "created_at": "2023-11-07T05:31:56Z"
}

Authorizations

Authorization
string
header
required

API Key authentication. Format: "Bearer YOUR_API_KEY"

Path Parameters

evaluation_id
string
required

Body

application/json
user_message
string
required

Your question or message about the evaluation results

Maximum string length: 5000
Example:

"Why did some samples score lower than others?"

chat_model
enum<string>
default:anthropic/claude-sonnet-4-5-thinking

AI model to use for chat (optional)

Available options:
anthropic/claude-sonnet-4-5,
anthropic/claude-sonnet-4-5-thinking,
anthropic/claude-haiku-4-5

Response

Chat response received

evaluation_id
string<uuid>
required
user_message
string
required

User's question/query

Minimum string length: 1
assistant_message
string
required

Assistant's response

Minimum string length: 1
id
string<uuid>
user
integer
user_email
string
created_at
string<date-time>