Qwen3.8 Max API

Trillion-scale MoE flagship for coding, long-horizon agents, and professional work, with image and video understanding across a 1M-token context.

Alibaba Cloud Text Generation 983.6K context Released Aug 3, 2026 Singapore Proprietary Endpoint

About Qwen3.8 Max

Trillion-scale MoE flagship for coding, long-horizon agents, and professional work, with image and video understanding across a 1M-token context.

Supports text, image, and video input, thinking mode with enable_thinking and thinking_budget up to 262144 tokens, function calling, structured output including strict JSON Schema, and five built-in tools: tool_web_search, tool_web_extractor, tool_code_interpreter, tool_web_search_image, and tool_image_search. Thinking tokens are billed as output tokens.

Also known as Alibaba Cloud Qwen3.8 Max

reasoning vision multimodal web search code interpreter function calling structured output agentic coding

Qwen3.8 Max specs

Model ID
qwen3-8-max
Author
Alibaba Cloud
Category
Text Generation
Released
Aug 3, 2026
Context window
983.6K tokens
Max output
131,072 tokens
Input
TextImageVideo
Output
Text
Structured output
JSON Schema
Region
Singapore
Endpoints
POST/v1/chat/completionsPOST/v1/responsesPOST/v1/messagesPOST/v1beta/models/qwen3-8-max:generateContent
Alternate model IDs
qwen3.8-max

Qwen3.8 Max API pricing

Live pay-as-you-go rates from the EmpirioLabs catalog. You can also choose a plan with weekly allowances.

Type
Spec
Rate
Input
per 1M prompt tokens
$2.00
Output
per 1M generated tokens
$6.00
Web search
per call when invoked
$0.02
Text-to-image search
per call when invoked
$0.02
Image-to-image search
per call when invoked
$0.02
Web extractor
per call when invoked
$0.00
Code interpreter
per call when invoked
$0.00
Compare on the full pricing page

How to call the Qwen3.8 Max API

Qwen3.8 Max serves the OpenAI-compatible Chat Completions API. Point any OpenAI SDK at https://api.empiriolabs.ai/v1 with your EmpirioLabs API key and use the model id qwen3-8-max. Get an API key from the EmpirioLabs dashboard.

cURL
curl https://api.empiriolabs.ai/v1/chat/completions \
  -H "Authorization: Bearer $EMPIRIOLABS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "qwen3-8-max",
    "messages": [
      {"role": "user", "content": "Write a haiku about the ocean."}
    ]
  }'
Python (OpenAI SDK)
from openai import OpenAI

client = OpenAI(
    base_url="https://api.empiriolabs.ai/v1",
    api_key="YOUR_EMPIRIOLABS_API_KEY",
)

response = client.chat.completions.create(
    model="qwen3-8-max",
    messages=[{"role": "user", "content": "Write a haiku about the ocean."}],
)
print(response.choices[0].message.content)
Full Qwen3.8 Max API reference

Qwen3.8 Max API parameters

Request parameters supported by the Qwen3.8 Max API on EmpirioLabs. Defaults apply when a field is omitted.

ParameterTypeDefaultRange / valuesDescription
temperaturenumber0.70 to 1.99Sampling temperature. 0 is deterministic and 1.99 is maximum randomness.
top_pnumber0.90 to 1Nucleus sampling probability mass. Lower values make outputs more focused.
max_tokensnumber-1 to 131072Maximum output tokens. When omitted, the model's own default applies.
stopstring--Up to 4 strings where the model will stop generating further tokens.
enable_thinkingbooleantrue-Enable reasoning before answering.
reasoning_effortenum-none, low, medium, high, maxReasoning effort level. none disables thinking. low, medium, high, and max set bounded thinking budgets sized to the selected model.
thinking_budgetnumber-1 to 262144Maximum tokens reserved for reasoning when thinking is enabled. When omitted, the model decides how long to reason.
preserve_thinkingbooleantrue-Include the reasoning of earlier assistant turns in the model's input on /v1/chat/completions. Return each turn's reasoning unchanged in that assistant message's...
seedinteger-0 to 2147483647Random seed for more reproducible sampling. Reusing a seed with the same request returns similar, not always identical, output.
top_kinteger-1 to 100Sample only from the K most likely tokens. When omitted, the model's own default applies.
presence_penaltynumber--2 to 2Penalty for tokens that already appeared in the generated text. Positive values encourage new topics. When omitted, the model's own default applies.
repetition_penaltynumber-0.1 to 2Higher values discourage repeated phrasing, and 1 applies no penalty. When omitted, the model's own default applies. Not available when a built-in tool is on, or on...
ninteger11 to 4Number of answers to return. Above 1 runs only on /v1/chat/completions, with enable_thinking false and without tools, web search or logprobs. The output tokens of...
logprobsbooleanfalse-Return the log probabilities of the answer tokens, not the reasoning, on /v1/chat/completions. Not available when a built-in tool is on.
13 more parameters in the docs

Good to know

Text, image, and video input are supported. Web search, web extractor, code interpreter, text-to-image search, and image-to-image search are optional built-in tools exposed through tool_* parameters. Web search, text-to-image search, and image-to-image search add $0.02 for each invoked call; web extractor and code interpreter run at no extra cost. Web extractor requires web search, and both web extractor and code interpreter require thinking. A single request can invoke a tool more than once, and each invoked call is billed. Thinking tokens are billed as output tokens. The qwen3-8-max id always serves the current Qwen3.8 Max build and moves with Alibaba's snapshot upgrades; since September 5, 2026 it serves the qwen3-8-max-0902 snapshot. Use qwen3-8-max-0902 when output must stay pinned to that exact snapshot.

Per-tool billing (usage.tool_usage)

When this model invokes built-in tools inside a single request, the response carries a normalized usage.tool_usage map alongside the token counts:

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

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

:variant1

Text, image, and video input are supported. Web search, web extractor, code interpreter, text-to-image search, and image-to-image search are optional built-in tools exposed through tool_* parameters. Web search, text-to-image search, and image-to-image search add $0.02 for each invoked call; web extractor and code interpreter run at no extra cost. Web extractor requires web search, and both web extractor and code interpreter require thinking. A single request can invoke a tool more than once, and each invoked call is billed. Thinking tokens are billed as output tokens. This id always serves the current Qwen3.8 Max build on the China endpoint and moves with Alibaba's snapshot upgrades. Use qwen3-8-max-0902 when output must stay pinned to the September 2, 2026 snapshot.

Qwen3.8 Max variants

Variants are alternate versions of Qwen3.8 Max with their own model id. Depending on the variant, they can differ in serving region, pricing, or supported parameters; everything else works the same way.

Qwen3.8 Max :variant1Save up to 18%

Model ID
qwen3-8-max:variant1
Region
China
Context window
983.6K tokens
Structured output
JSON Schema
Batch API
Available, 35% off list price
Alternate model IDs
qwen3.8-max:variant1
Type
Spec
Rate
Input
per 1M prompt tokens
$2.00$1.65
Output
per 1M generated tokens
$6.00$4.951
Web search
per call when invoked
$0.01
Text-to-image search
per call when invoked
$0.01
Image-to-image search
per call when invoked
$0.01
Web extractor
per call when invoked
$0.00
Code interpreter
per call when invoked
$0.00
Try :variant1 in the playground

Qwen3.8 Max on the blog

Announcements, prompt tests, and comparisons that cover this model.

Qwen3.8 Max API: common questions

How much does the Qwen3.8 Max API cost?

On EmpirioLabs, Qwen3.8 Max is billed pay as you go: Input $2.00 per 1M prompt tokens; Output $6.00 per 1M generated tokens; Web search $0.02 per call when invoked; Text-to-image search $0.02 per call when invoked; Image-to-image search $0.02 per call when invoked; Web extractor $0.00 per call when invoked; Code interpreter $0.00 per call when invoked. You can also choose a plan with weekly allowances. The live rate card on this page always matches what the API charges.

What inputs and outputs does Qwen3.8 Max support?

Qwen3.8 Max takes Text, Image, Video as input and returns Text.

What is the context window of Qwen3.8 Max?

Qwen3.8 Max supports a 983.6K-token context window with up to 131,072 output tokens per response.

Is the Qwen3.8 Max API OpenAI-compatible?

Yes. Qwen3.8 Max serves the OpenAI-compatible Chat Completions API, so existing OpenAI SDKs work by pointing base_url at https://api.empiriolabs.ai/v1 and setting the model id to qwen3-8-max.

What is the model ID for Qwen3.8 Max?

Send qwen3-8-max as the model in your request. The alternate IDs qwen3.8-max also work.

When was Qwen3.8 Max released?

Alibaba Cloud released Qwen3.8 Max on Aug 3, 2026.

Where is Qwen3.8 Max served from?

Qwen3.8 Max on EmpirioLabs is served from Singapore.

Which Qwen3.8 Max variants are available?

Qwen3.8 Max is available as 2 model ids: the default qwen3-8-max plus qwen3-8-max:variant1 (China). Variants can differ in serving region, pricing, or supported parameters; the rate cards for each are on this page.

Can I try Qwen3.8 Max in the browser before integrating?

Yes. The EmpirioLabs playground runs Qwen3.8 Max in the browser with the same parameters the API exposes, so you can test prompts before writing code.

How do I get a Qwen3.8 Max API key?

Create an EmpirioLabs account, then generate a key under API Keys in the dashboard. Billing is pay-as-you-go credits, so you only pay for the requests you make. You can also choose a plan with weekly allowances.

Compare Qwen3.8 Max with other models

All model comparisons

Ready to use better endpoints?

Check out our pricing or reach out if you want your own model deployed on our stack.