Qwen3.8-Max - AnyFast

Qwen3.8-Max is Alibaba’s flagship Qwen3.8 multimodal model, available through AnyFast’s OpenAI-compatible Chat Completions API at POST /v1/chat/completions. Use the model ID qwen3.8-max for text generation, image and video understanding, function calling, and structured output.

Key capabilities

1M-token context window — The upstream specification allows 1,000,000 tokens.
131,072-token maximum output — Upstream limit; available output also depends on context.
Multimodal input — Accept text, images, and video.
Text output — Return text responses.
Thinking mode — Enabled by default and can be disabled.

Quick example

cURL

curl --request POST \
  --url https://www.anyfast.ai/v1/chat/completions \
  --header 'Authorization: Bearer YOUR_API_KEY' \
  --header 'Content-Type: application/json' \
  --data '{
  "model": "qwen3.8-max",
  "messages": [
    {
      "role": "user",
      "content": "Explain a solar eclipse in one sentence."
    }
  ],
  "max_completion_tokens": 1024,
  "enable_thinking": false
}'

Python

from openai import OpenAI

client = OpenAI(api_key="YOUR_API_KEY", base_url="https://www.anyfast.ai/v1")
response = client.chat.completions.create(
    model="qwen3.8-max",
    messages=[{"role": "user", "content": "Explain a solar eclipse in one sentence."}],
    max_completion_tokens=1024,
    extra_body={"enable_thinking": False},
)
print(response.choices[0].message.content)

Image and video input

Send image content as an image_url part. Image and video content parts were both successfully called through AnyFast for qwen3.8-max. Use a public HTTPS URL or a supported data URI; see the visual understanding guide for upstream media limits.

Python

response = client.chat.completions.create(
    model="qwen3.8-max",
    messages=[{
        "role": "user",
        "content": [
            {"type": "text", "text": "Describe this image."},
            {"type": "image_url", "image_url": {"url": "https://example.com/photo.jpg"}},
        ],
    }],
    max_completion_tokens=300,
    extra_body={"enable_thinking": False},
)
print(response.choices[0].message.content)

For a video, replace the image_url part with {"type": "video_url", "video_url": {"url": "https://example.com/sample.mp4"}}. The video URL must point to an accessible file.

Thinking and structured output

Thinking is enabled by default. Set enable_thinking to false to omit the reasoning step. In the OpenAI SDK, pass this provider-specific field through extra_body. A non-thinking json_object request succeeded through AnyFast for qwen3.8-max.

Python

response = client.chat.completions.create(
    model="qwen3.8-max",
    messages=[{"role": "user", "content": "Return a JSON object with an answer key."}],
    response_format={"type": "json_object"},
    extra_body={"enable_thinking": False},
)
print(response.choices[0].message.content)

Function calling

Define functions in tools. When the model returns finish_reason: "tool_calls" and message.tool_calls, execute the function in your application and continue the conversation. This request shape succeeded through AnyFast. Use tool_choice: "required" to require a tool call.

Python

tools = [{
    "type": "function",
    "function": {
        "name": "get_weather",
        "description": "Get weather for a city",
        "parameters": {
            "type": "object",
            "properties": {"city": {"type": "string"}},
            "required": ["city"],
        },
    },
}]
response = client.chat.completions.create(
    model="qwen3.8-max",
    messages=[{"role": "user", "content": "Call get_weather for Singapore."}],
    tools=tools,
    tool_choice="required",
    extra_body={"enable_thinking": False},
)
print(response.choices[0].message.tool_calls)

Streaming

Set stream=True to receive SSE chunks. The verified final usage-only chunk has usage and an empty choices array.

Python

stream = client.chat.completions.create(
    model="qwen3.8-max",
    messages=[{"role": "user", "content": "Say hello briefly."}],
    stream=True,
    stream_options={"include_usage": True},
    extra_body={"enable_thinking": False},
)
for chunk in stream:
    if chunk.choices:
        print(chunk.choices[0].delta.content or "", end="")
    if chunk.usage:
        print("\nTotal tokens:", chunk.usage.total_tokens)

Parameters

Parameter	Type	Required	Description
`model`	string	Yes	Exact ID: `qwen3.8-max`.
`messages`	array	Yes	Conversation messages; text, image, or video content parts.
`max_completion_tokens`	integer	No	Maximum generated tokens; upstream limit is 131,072.
`enable_thinking`	boolean	No	Defaults to true; false disables thinking.
`stream`	boolean	No	Return Server-Sent Events when true.
`stream_options`	object	No	Set `include_usage: true` to receive usage in the final stream chunk.
`tools`	array	No	Functions available to the model.
`tool_choice`	string	No	Verified with `required`; `auto` and `none` are also available.
`response_format`	object	No	`json_object` was verified with thinking disabled.

API Reference

Request and response fields for Qwen3.8-Max.

​Key capabilities

​Quick example

​Image and video input

​Thinking and structured output

​Function calling

​Streaming

​Parameters

API Reference

​Official references

Key capabilities

Quick example

Image and video input

Thinking and structured output

Function calling

Streaming

Parameters

Official references