Bulk create dataset logs

curl -X POST https://api.respan.ai/api/datasets/dataset_id/logs/bulk/ \
     -H "Authorization: Bearer sk_live_xxxxx" \
     -H "Content-Type: application/json" \
     -d '{
  "logs": [
    {
      "input": "What is your return policy?",
      "expected_output": "You can return within 30 days",
      "metadata": {
        "category": "support"
      }
    },
    {
      "input": [
        {
          "role": "user",
          "content": "Hello"
        }
      ],
      "output": {
        "role": "assistant",
        "content": "Hi there!"
      },
      "metrics": {
        "tokens": 10,
        "cost": 0.0001
      }
    }
  ]
}'

{
  "success_count": 95,
  "error_count": 5,
  "errors": [
    {
      "index": 3,
      "error": "Invalid input format"
    }
  ],
  "total_errors": 5
}

Create multiple dataset logs in one request from an array of unified-format log objects. Partial success is allowed.

Path parameters

dataset_idstringRequired

Dataset ID. Use _saved_logs for the virtual saved-logs collection.

Request

This endpoint expects an object.

logslist of objectsRequired

Response

Bulk create completed. Some rows may still contain errors.

success_countinteger

error_countinteger

errorslist of objects

total_errorsinteger

Errors

401

Unauthorized Error

Bulk create dataset logs

Bulk create dataset logs

Path parameters

Headers

Request

Response

Errors

Path parameters

Headers

Request

Response

Errors

$	curl -X POST https://api.respan.ai/api/datasets/dataset_id/logs/bulk/ \
>	-H "Authorization: Bearer sk_live_xxxxx" \
>	-H "Content-Type: application/json" \
>	-d '{
>	"logs": [
>	{
>	"input": "What is your return policy?",
>	"expected_output": "You can return within 30 days",
>	"metadata": {
>	"category": "support"
>	}
>	},
>	{
>	"input": [
>	{
>	"role": "user",
>	"content": "Hello"
>	}
>	],
>	"output": {
>	"role": "assistant",
>	"content": "Hi there!"
>	},
>	"metrics": {
>	"tokens": 10,
>	"cost": 0.0001
>	}
>	}
>	]
>	}'

1	{
2	"success_count": 95,
3	"error_count": 5,
4	"errors": [
5	{
6	"index": 3,
7	"error": "Invalid input format"
8	}
9	],
10	"total_errors": 5
11	}