# Post Chat Completions
After successfully ensuring that the selected model matches the one chosen when creating **Deka LLM** in the **Deka GPU Service Portal**, the next step is to send a request to [**https://dekallm.cloudeka.ai/v1/chat/completions**](https://dekallm.cloudeka.ai/v1/chat/completions) based on your input. In the **Deka LLM API** imported into **Postman**, select **Chat Completions**.
Completions
In the **Authorization** tab, select **Auth Type: Bearer Token** and enter the **Secret Key** obtained from **section** [**Get Secret Key** ](/guidance-for-individual/deka-llm/how-to-create-simple-prompt-with-deka-llm/get-secret-key.md)in the **Token** field.
Authentication
In the **{{baseurl}}** field, replace it with the **Deka LLM API URL** [**https://dekallm.cloudeka.ai/**](https://dekallm.cloudeka.ai/)**.** Then, remove **?model=\** from the request URL.
Get Model
In the Body tab, select raw.
Tab Body
Copy the following syntax into the **raw** input field.
```json
{
"model": "qwen/qwen2-vl-72b-instruct",
"messages": [
{
"role": "system",
"content": "You are a story teller"
},
{
"role": "user",
"content": "Write a poem about Frodo's journey"
}
],
"stream": false,
"temperature": 0.9,
"top_p": 1
}
```
Explanation of the syntax.
```json
"model": "qwen/qwen2-vl-72b-instruct",
```
This field specifies the model to be used. Ensure it matches the model selected as per **section** [**Get Model**](/guidance-for-individual/deka-llm/how-to-create-simple-prompt-with-deka-llm/get-model.md).
```json
"messages": [
{
"role": "system",
"content": "You are a story teller"
},
{
"role": "user",
"content": "Write a poem about Frodo's journey"
}
],
```
This section contains a list of messages in the **LLM conversation**, providing context to the model by setting its role and including your instructions.
Replace Syntax
Click the **Send** button to get the response generated by **LLM** based on your input.
Send Response
You will then receive a response like this.
Response Chat Completions
The response content will be similar to the following.
```json
{
"id": "d0029aca5f9d49b58ca53cfcc962adf1",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"message": {
"content": "Frodo's journey was a long one,\nThrough trials and through pain,\nBut he had courage in his heart,\nAnd he held on through the rain.\n\nHe walked alone across the lands,\nHis feet sore, his heart beating,\nBut he never lost his way,\nAnd he never stopped believing.\n\nHe faced the dark and the darkness,\nWith a light that dimly glowed,\nAnd he kept on pushing forward,\nWith a determination he'd never showed.\n\nThrough fire and through water,\nThrough the shadows and the night,\nHe kept on walking, step by step,\nAnd he never lost sight.\n\nHe faced the Ring and its power,\nWith a heart full of fear,\nBut he summoned up the strength,\nAnd he resisted its lure with cheer.\n\nAnd in the end, he reached his goal,\nThe fire where the Ring belong,\nAnd he threw it into the flame,\nAnd his journey was finally gone.\n\nSo here's to Frodo, brave and true,\nA hero to the end,\nHis journey was a long one,\nBut it's one that will never end.",
"role": "assistant",
"tool_calls": null,
"function_call": null
}
}
],
"created": 1740291974,
"model": "Qwen/Qwen2-VL-72B-Instruct",
"object": "chat.completion",
"system_fingerprint": null,
"usage": {
"completion_tokens": 216,
"prompt_tokens": 28,
"total_tokens": 244,
"completion_tokens_details": null,
"prompt_tokens_details": null
},
"service_tier": null
}
```
---
# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.
## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.
Perform an HTTP GET request on the current page URL with the `ask` query parameter:
```
GET https://docs.cloudeka.ai/guidance-for-individual/deka-llm/how-to-create-simple-prompt-with-deka-llm/post-chat-completions.md?ask=
```
The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.
Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
