Qwen Series

Qwen 3.7-generation flagship model built for the agent era, with comprehensive improvements in coding, office work, and long-horizon autonomous execution. Supports thinking mode toggle, function calling, and web search. 1M context.

Cost-effective Plus model in the Qwen3.7 series. On top of strong text capabilities, it comprehensively upgrades vision-language abilities: it can read the screen and operate a GUI, generate code from visual references, and retain full coding, tool-use, and productivity-workflow agent capabilities. 1M context.

The most powerful preview model in the Qwen3.6 series, ideal for complex reasoning, multi-step code generation, and tool-based tasks.

Balanced flagship model supporting a 1M-token context window, function calling, and built-in tools, ideal for most production scenarios.

Balanced performance model, ideal for most production scenarios. Strong language ability and fast responses.

High-speed response model, ideal for latency-sensitive scenarios. Highly cost-effective.

Flagship omni-modal model supporting any combination of text, image, audio, and video input, with text and speech output. 113 input languages, 55 voices, with web search and voice cloning.

Lightweight omni-modal model supporting any combination of text, image, audio, and video input with text + speech output. A cost-effective omni-modal choice.

from openai import OpenAI

client = OpenAI(
    api_key="sk-air-your-key",
    base_url="https://nexusflow.hk/v1",
)

# Use Qwen to process long text
response = client.chat.completions.create(
    model="qwen3.6-plus",
    messages=[
        {"role": "system", "content": "You are a professional document analysis assistant."},
        {"role": "user", "content": "Summarize the main points of the following long document... (you can paste very long text here)"}
    ],
    max_tokens=4096,
)

print(response.choices[0].message.content)