Chat Completion with Qwen3.8 Flash 27B

What it does

Chat Completion generates a text response from a chat-style message list. Send a model and a messages array with one or more entries, each with a role of system, user, or assistant and a content string. You can also set stream to true for streamed output, and max_tokens to control the length of the reply.

Use Chat Completion when you need a conversational LLM endpoint for assistants, drafting, classification, or controlled text generation inside your app. The request format is simple and keeps the conversation history explicit, so you can build prompts from your own application state and send exactly the context you want the model to use.

The response schema is open-ended, so the exact payload depends on the model and runtime. In practice, you should treat the result as the completion object returned by the model call and parse it according to your integration requirements. This makes Chat Completion suitable for workflows where you want to route user input to an LLM and handle the generated text in your own code.

POST

Generate a Qwen text chat completion

https://api.eu.apyhub.com/flonno/chat-completion-with-qwen3-8-flash-27b/chat/completions

QUICKSTART

GUIDE

Quickstart

Create a chat completion with the required model and messages.

curl -X POST "https://api.eu.apyhub.com/flonno/chat-completion-with-qwen3-8-flash-27b/chat/completions" \
  -H "apy-token: $APY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "qwen3.8-27b",
    "messages": [
      {
        "role": "user",
        "content": "Hello, how are you?"
      }
    ],
    "max_tokens": 256
  }'

What you'll get back

A JSON object. The generated text is in choices[0].message.content.

{
  "id": "chatcmpl-3e13a3a3f428b2a7",
  "object": "chat.completion",
  "created": 1791873923,
  "model": "qwen3.8-27b",
  "choices": [
    {
      "finish_reason": "stop",
      "index": 0,
      "logprobs": null,
      "message": {
        "content": "I'm doing well, thank you for asking! How are you doing today? Is there anything specific I can help you with?",
        "refusal": null,
        "role": "assistant"
      }
    }
  ],
  "usage": {
    "completion_tokens": 27,
    "completion_tokens_details": {
      "reasoning_tokens": 0
    },
    "prompt_tokens": 15,
    "prompt_tokens_details": {
      "cached_tokens": 0
    },
    "total_tokens": 42
  }
}

TRY ITLIVE · 1 ATOM

API key

Loading your default key…

The full key is used to call the gateway and stays in this tab — never sent to orbit or saved.

About this endpoint

What it does

Generates a text chat completion from the messages you provide. You send a model name and an array of chat messages, and the endpoint returns the model's reply along with token usage.

Request Body

Parameter	Type	Mandatory	Description
model	String	Yes	Model identifier to use for the completion, e.g. `qwen3.8-27b`.
messages	Object Array	Yes	Chat message list. Must contain at least 1 item.
messages[].role	ENUM	Yes	Message role. Allowed values: `system`, `user`, `assistant`.
messages[].content	String	Yes	Message text content.
max_tokens	Integer	No	Maximum number of tokens to generate. Default: `128`. Minimum: `1`.
stream	Boolean	No	Whether to stream the response. Default: `false`.

Additional model-specific parameters are accepted and passed through to the model.

Response

Returns a JSON object. The reply text is in choices[0].message.content.

Field	Type	Description
id	String	Unique identifier for the completion.
object	String	Object type, always `chat.completion`.
created	Integer	Unix timestamp (seconds) when the completion was created.
model	String	Model that generated the completion.
choices	Object Array	List of generated completions.
choices[].index	Integer	Position of the choice in the list.
choices[].finish_reason	String	Why generation stopped: `stop` (natural end) or `length` (hit `max_tokens`).
choices[].logprobs	Object	Log probability data. `null` unless the model returns it.
choices[].message.role	String	Always `assistant`.
choices[].message.content	String	The generated text.
choices[].message.refusal	String	Refusal message if the model declined to answer, otherwise `null`.
usage.prompt_tokens	Integer	Tokens in the input messages.
usage.completion_tokens	Integer	Tokens in the generated reply.
usage.total_tokens	Integer	Sum of prompt and completion tokens.
usage.completion_tokens_details.reasoning_tokens	Integer	Tokens used for internal reasoning, `0` for non-reasoning models.
usage.prompt_tokens_details.cached_tokens	Integer	Prompt tokens served from cache.

The response may include extra fields depending on the model.

Notes

max_tokens defaults to 128, so longer replies are cut off with finish_reason: "length". Increase it for longer output.
Include earlier user and assistant turns in messages to keep conversation context. The endpoint does not store history between calls.
With stream: true, the reply is returned incrementally instead of as one JSON object.

▣ COMMON ERRORS

Errors any endpoint can return

400bad_request

Required parameter missing or malformed body.

401unauthorized

API key missing, revoked, or not authorized for this service.

429rate_limited

Your plan's per-second rate exceeded. Retry with exponential backoff.

503upstream_busy

Backend temporarily unavailable. Try again in a few seconds.

Similar APIs

Closest matches in the catalog based on this service.

CALLChat Completion Generation APIDosvak LLC CALLChat Completion with Qwen3-Next-80BMasterA AI CALLSimple Chat Prompt APIDosvak LLC CALLSummarize Text APISharpAPI

From the blog

Reads that touch this API.

How to Extend ChatGPT's Capabilities With Plugins: Files, Data, Validation and MoreSep 22, 2026 Enable Voice-based Interactions in your Retool ApplicationAug 7, 2026 ChatGPT, Claude and Grok Went Down Together. What That Says About How We Use Them.Sep 8, 2026 Best Free AI Chatbots You Can Use in 2026Aug 10, 2026