Qwen3.6 Max Preview

POST /v1/chat/completions

Largest preview variant in the 3.6 series (text-only): improved coding agent execution, stronger front-end skills, and broader long-tail knowledge.

This model is deprecated and will be retired on 2026-09-08. After that date, requests to this model will fail. Migrate to a successor model before then.

At a glance

Field	Value
Model id	`qwen3-6-max-preview`
Input modalities	Text
Output modalities	Text
Context window	256K
Weight precision	-
Max output tokens	65,536
Region	Singapore
Features	reasoning, agentic_coding, web_search
Native inference	No
New	Yes
Supported endpoints	`POST /v1/chat/completions`, `POST /v1/responses`, `POST /v1/messages`
Deprecation date	2026-09-08

Pricing

Charge	Spec	Rate
Input	per 1M prompt tokens	<=128K $1.31; 128K-256K $1.97
Output	per 1M generated tokens	<=128K $7.88; 128K-256K $11.82
Web Search	per call	$0.020

Example request

$ curl https://api.empiriolabs.ai/v1/chat/completions \
>   -H 'Authorization: Bearer $EMPIRIOLABS_API_KEY' \
>   -H 'Content-Type: application/json' \
>   -d '{"model": "qwen3-6-max-preview", "messages": [{"role":"user","content":"Hello"}]}'

Parameters

Parameter	Type	Required	Default	Description
`temperature`	number	no	`0.7`	Sampling temperature · Range: 0 – 2
`top_p`	number	no	`1.0`	Nucleus sampling · Range: 0 – 1
`max_tokens`	number	no	`4096`	Max output tokens · Range: 1 – 65536
`frequency_penalty`	number	no	`0`	Penalty for repeated tokens. >0 reduces repetition, <0 encourages it. · Range: -2 – 2
`presence_penalty`	number	no	`0`	Penalty for new vs. seen tokens. >0 encourages new topics, <0 encourages staying on topic. · Range: -2 – 2
`reasoning_effort`	enum	no	`"medium"`	Reasoning effort level. none disables thinking. low, medium, high, and max set bounded thinking budgets sized to the selected model. Sent as an OpenAI-style reasoning_effort field, translated into enable_thinking and thinking_budget for the model service. · Allowed: `none`, `low`, `medium`, `high`, `max`
`stop`	string	no	-	Comma-separated stop sequences
`enable_thinking`	boolean	no	true	Reason step-by-step before answering
`thinking_budget`	number	no	`32768`	Tokens reserved for thinking · Range: 1 – 393216
`tool_web_search`	boolean	no	false	Search the web for real-time information.
`disable_formatting`	boolean	no	false	Skip the EmpirioLabs Markdown formatting (citation [N] rewriting + References block when web search / tools were used). The raw upstream answer with plain [N] citations is returned.

Notes

Pricing is ~1.5x above 128K tokens. Plain-text capabilities only in this preview; multimodal not yet enabled.

Per-tool billing (usage.tool_usage)

When this model invokes tools (web search, code interpreter, etc.) inside a single request, the response carries a normalized usage.tool_usage map alongside the token counts. The example below shows the shape — exact field names, units, and which tools appear can vary slightly per provider:

1 "usage": {
2   "prompt_tokens": 123,
3   "completion_tokens": 456,
4   "cost_usd": 0.0042,
5   "tool_usage": {"web_search": 3, "code_interpreter": 1}
6 }

The tool counts are already factored into cost_usd — they are surfaced for transparency so you can audit per-tool billing. The field is omitted when no tools were invoked.

Machine-readable schema: GET https://api.empiriolabs.ai/v1/models/qwen3-6-max-preview.