Grok Image Editing: What Is the Image ID Format?

Short answer: Grok image editing input format

There is no separate image ID in a Grok image edit request. LLMTR (llmtr.com) is a gateway that gives access to language models hosted in Türkiye and at global providers through one OpenAI-compatible API. Use https://llmtr.com/v1 and POST /v1/images/edits with xai/grok-imagine-image; the image field accepts an HTTPS URL or data:image/ value.

On LLMTR, xAI file_id is not accepted. Use image for one input or images for 1-5 inputs. A local file can be sent as a base64 data:image/ URI. With response_format b64_json, store returned bytes; response URLs are short-lived, so download a URL immediately if you choose url.

What is the image id format in Grok image editing?

Grok image editing uses the image field for the image or a link, not a separate image ID. On xAI, the hosted-file form is file_id; on LLMTR, an edit can contain 1 image in image or 1-5 reference images in images.

LLMTR accepts an HTTPS URL string, a URL object, an image_url object, or a base64 data URI. xAI also documents file_id from the xAI Files API, but LLMTR does not accept file_id. On LLMTR, send the image itself or its URL in the accepted image field.

• xAI docs, Image editing · 2026-10-06

The request on LLMTR: POST /v1/images/edits

Create an edit request with POST https://llmtr.com/v1/images/edits. The endpoint is available only for xai/grok-imagine-image and xai/grok-imagine-image-quality. Send image for one image or images for an array of 1-5 reference images, and use the required prompt field to state what to change.

Choose response_format as url or b64_json. A response URL is short-lived, so use b64_json and store the returned bytes when the result must be kept, or download the URL immediately. The optional resolution field accepts 1k or 2k for the edit request.

Working example with a base64 data URI and b64_json

Build a local image as a data URI by joining "data:image/png;base64," with the base64 of the file bytes. Send that URI in the image field of a POST request to https://llmtr.com/v1/images/edits with xai/grok-imagine-image and response_format set to b64_json for the edit request.

The example below reads product.png, creates a data:image/png;base64, value, posts the request, decodes b64_json, and writes edited.png or edited.jpg. It uses xai/grok-imagine-image and an LLMTR_API_KEY authorization header. Saving decoded bytes avoids relying on a short-lived response URL for the returned result.

Python example that sends product.png as a base64 data URI and saves the returned b64_json image.

import base64
import os
import requests

with open("product.png", "rb") as f:
    data_uri = "data:image/png;base64," + base64.b64encode(f.read()).decode()

response = requests.post(
    "https://llmtr.com/v1/images/edits",
    headers={"Authorization": f"Bearer {os.environ['LLMTR_API_KEY']}"},
    json={
        "model": "xai/grok-imagine-image",
        "prompt": "Keep the product, replace the background with a clean white studio",
        "image": {"type": "image_url", "url": data_uri},
        "response_format": "b64_json",
    },
    timeout=180,
)
response.raise_for_status()
image_bytes = base64.b64decode(response.json()["data"][0]["b64_json"])
extension = "png" if image_bytes[:4] == b"\x89PNG" else "jpg"
with open(f"edited.{extension}", "wb") as out:
    out.write(image_bytes)

Errors and cost

Only xai/grok-imagine-image and xai/grok-imagine-image-quality are available on the LLMTR edit endpoint. Calling /v1/images/edits with a Google or OpenAI model returns 400 unsupported_operation; Google reference-image generation runs on /v1/images/generations instead. A model outside those two cannot be used for this edit operation.

On LLMTR, xai/grok-imagine-image costs $0.02 per image and xai/grok-imagine-image-quality costs $0.05 per image. Edit requests count both the images sent and the images produced. Billing follows the actual cost when xAI reports one; otherwise, the per-image price applies for the request.

• xAI docs, Pricing · 2026-10-06

Getting started on LLMTR: account, credit, key and first request

Step 1: create an account at llmtr.com and verify your email. Step 2: top up credit from the Dashboard through secure checkout. An 8% platform margin is added once, on top of the requested top-up amount: a $10.00 credit is charged $10.80. The platform margin is not added to model prices and every API call is deducted at the model's catalog price.

Step 3: create an API key under Dashboard > API Keys. The raw key is shown only once; store it as an environment variable such as LLMTR_API_KEY. Step 4: send the first request to https://llmtr.com/v1 with any OpenAI-compatible SDK by changing only base_url and api_key. Usage and spend per key are visible on the Dashboard usage page. The same key and same credit balance work for every model in the catalog; switching models changes the model string.

Set up LLMTR access and send a Grok image edit request

Create an account, add credit, create an API key, and send a Grok image edit request.

1. Create an account. Create an account at llmtr.com and verify your email.
2. Top up credit. Top up credit from the Dashboard through the secure checkout. An 8% platform margin is added once, on top of the requested top-up amount; for example, $10.00 credit is charged $10.80. The margin is not added to model prices, and every API call is deducted at the model's catalog price.
3. Create an API key. Create an API key under Dashboard > API Keys. The raw key is shown only once, so store it as an environment variable such as LLMTR_API_KEY.
4. Send the first request. Send the first request to https://llmtr.com/v1/images/edits with any OpenAI-compatible SDK by changing only base_url and api_key. Usage and spend per key are visible on the Dashboard usage page. The same key and same credit balance work for every model in the catalog; switching models means changing the model string.

Frequently asked questions

What format does the image field use in Grok image editing?

On xAI, use an HTTPS URL, a base64 data URI, or a file_id from the xAI Files API. On LLMTR, use image for 1 input or images for 1-5 inputs, with an HTTPS URL string, URL object, image_url object, or base64 data URI; file_id is not accepted.

How many reference images can a Grok edit request take on LLMTR?

Use image for one image or images for an array of 1-5 reference images. Both forms are accepted, and one of them is required.

How much does a Grok image edit cost?

xai/grok-imagine-image costs $0.02 per image; xai/grok-imagine-image-quality costs $0.05 per image. Images sent and produced both count, and billing follows xAI's actual cost when reported, otherwise the per-image price applies.

Related posts

• xAI API Images: aspect_ratio, Grok Imagine and b64_json
• Grok Imagine Image: why the standard tier is enough for most product work

• Image generation guide
• Model catalog
• Pricing