OpenLuxQwen models API
Qwen3 Max delivers dependable Chinese and coding performance at $0.0784 per million tokens on OpenLux — 88% below official list pricing — with tool calling on the standard OpenAI-compatible endpoint.
Calling Qwen models on OpenLux
Qwen models are widely used in Chinese-language products and in engineering workflows where instruction following and structured output matter. Qwen3 Max handles chat, code and tool use in one model, which keeps a pipeline simple: fewer models, fewer prompt variants, fewer surprises.
It is a natural pair with DeepSeek in bilingual stacks — roughly the same price bracket, with different strengths. Qwen leans toward coding and structured output; DeepSeek leans toward reasoning depth. Testing both on real traffic costs very little because they share the same key.
On OpenLux the rate is $0.0784 per million tokens, 88% below list, metered pay as you go with per-request cost shown in the console.
Why teams call Qwen on OpenLux
Chinese comprehension
Production-grade Chinese understanding and generation, including mixed Chinese-English input.
Coding and structure
Good at producing code and schema-shaped output that downstream pipelines can consume directly.
Tool calling
Function calling is supported, so it can act inside agents and workflow tools.
Predictable pricing
At $0.0784 per 1M tokens it is cheap enough to be a default rather than an exception.
Rates for Qwen models
| Model | Category | Rate | Discount |
|---|---|---|---|
| qwen3-max | Text | $0.0784 per 1M tokens | -88% |
Rates are USD, metered pay as you go with no subscription. See the full model catalogue and pricing plans for credit allowances.
Where Qwen models fit
Chinese-language products
Customer support, content and search experiences for Chinese-speaking users.
Data pipelines
Extraction and normalisation where reliable structured output matters more than creative writing.
Bilingual agents
Agents serving both languages, with tool calling for actions and code generation for tasks.
Qwen on OpenLux questions
How much does Qwen3 Max cost on OpenLux?
$0.0784 per 1 million tokens, 88% below official list pricing, metered pay as you go.
How do I call Qwen3 Max?
Set the base URL to the OpenLux endpoint and send "model": "qwen3-max" with your OpenLux key — the same pattern as every model in the catalogue.
Qwen3 Max or DeepSeek V4 — which should I use?
At $0.0784 vs $0.0613 per 1M tokens the prices are close: Qwen leans coding and structured output, DeepSeek leans reasoning depth. Both are on the same endpoint, so testing both on your own traffic costs nothing extra.
Does Qwen3 Max support tool calling?
Yes — function calling is supported for agent and pipeline integrations, along with streamed responses.
Start calling Qwen models today
One OpenLux key covers Qwen and every other provider in the catalogue, metered in USD with real-time cost in the console.