Skip to main content
POST
/
batches
# Docs for v1 can be found by changing the above selector ^ from together import Together import os client = Together( api_key=os.environ.get("TOGETHER_API_KEY"), ) batch = client.batches.create(input_file_id="file_id", endpoint="/v1/chat/completions") print(batch.job) 
{
  "job": {
    "id": "01234567-8901-2345-6789-012345678901",
    "user_id": "user_789xyz012",
    "input_file_id": "file-input123abc456def",
    "file_size_bytes": 1048576,
    "status": "IN_PROGRESS",
    "job_deadline": "2024-01-15T15:30:00Z",
    "created_at": "2024-01-15T14:30:00Z",
    "endpoint": "/v1/chat/completions",
    "progress": 75,
    "model_id": "Qwen/Qwen3.5-9B",
    "output_file_id": "file-output789xyz012ghi",
    "error_file_id": "file-errors456def789jkl",
    "error": "<string>",
    "completed_at": "2024-01-15T15:45:30Z"
  },
  "warning": "<string>"
}

Documentation Index

Fetch the complete documentation index at: https://docs.together.ai/llms.txt

Use this file to discover all available pages before exploring further.

Authorizations

​
Authorization
string
header
default:default
required

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

Body

application/json
​
endpoint
enum<string>
required

The endpoint to use for batch processing. Each line of the uploaded input file is dispatched against this endpoint.

  • /v1/chat/completions — chat completion batches
  • /v1/audio/transcriptions — audio transcription batches (e.g. openai/whisper-large-v3)
  • /v1/audio/translations — audio translation batches
Available options:
/v1/chat/completions,
/v1/audio/transcriptions,
/v1/audio/translations
Example:

"/v1/chat/completions"

​
input_file_id
string
required

ID of the uploaded input file containing batch requests

Example:

"file-abc123def456ghi789"

​
completion_window
string

Time window for batch completion (optional)

Example:

"24h"

​
priority
integer

Priority for batch processing (optional)

Example:

1

​
model_id
string

Model to use for processing batch requests

Example:

"Qwen/Qwen3.5-9B"

Response

Job created (potentially with warnings)

​
job
object
​
warning
string