Qwen4 Max

Qwen1M context

VisionReasoningTool callingWeb search
Official · Alibaba Cloud Model Studio
$1.50 / $4.50
per 1M in / outPricing page
Cheapest · DeepInfra
$1.41 / $4.24
−6%a month at your usage

As of Nov 17, 2026, Qwen4 Max costs $1.50 per million input tokens and $4.50 per million output tokens from Qwen. The cheapest of 5 providers is DeepInfra at $1.41 / $4.24. No price change since Appndo started tracking it on Nov 17, 2026. Recent price changes

Providers (5)

Sorted by your monthly estimate. Prices per million tokens. Alibaba Cloud Model Studio also lists up to 131K output tokens per request.

  • Input / output
    $1.41 / $4.24
    Context
    1M
    Per month
    $22.61

    Qwen/Qwen4-Max

  • Input / output
    $1.41 / $4.24
    Context
    1M
    Per month
    $22.61

    qwen/qwen4-max

  • Input / output
    $1.43 / $4.28
    Context
    1M
    Per month
    $22.80

    qwen/qwen4-max

  • Input / output
    $1.50 / $4.50
    Context
    992K
    Per month
    $24.00

    dashscope/qwen4-max

  • Input / output
    $1.88 / $5.63
    Context
    1M
    Per month
    $30.00

    qwen-4-max

Cloud and hosted providers may add regional pricing, minimums or platform fees. Supports: Vision, Reasoning, Tool calling, Web search.

Similar models

Common questions

How much does the Qwen4 Max API cost?

As of Nov 17, 2026, Qwen4 Max costs $1.50 per million input tokens and $4.50 per million output tokens at Qwen's official price.

Where is Qwen4 Max cheapest?

As of Nov 17, 2026, the cheapest of 5 providers is DeepInfra at $1.41 per million input tokens and $4.24 per million output tokens, about 6% less than the official price for a typical mix of 5 input tokens to 1 output token.

What is the context length of Qwen4 Max?

Qwen4 Max takes up to 992K tokens of context through Qwen's API; the limit differs between providers.

How does Qwen4 Max compare with similar models?

Qwen4 Max has no LMArena score yet. Other Qwen models and their cheapest price per million input / output tokens: Qwen 3 32B ($0.08 / $0.23), Qwen 3.8 27B ($0.17 / $0.50), and Qwen2.5 Coder 32B Instruct ($0.05 / $0.10).

About this data

Prices come from LiteLLM's model cost map, with more providers from models.dev (both MIT licence), checked Nov 17, 2026; confirm on the provider's pricing page before you rely on a number.