Cheaper Inference

A little less waiting. A lot more saving.

Getting your workspace ready…

qwen3-max-thinking API pricing, specs and examples · CheaperInference

Workloads to evaluate

Evaluate summarization, drafting, and question answering with examples from your own application.
Test tool calls against your application's actual function schemas before connecting actions.
Compare the available reasoning settings on tasks where additional deliberation is useful.

Limits and tradeoffs

The catalog does not list image or video input for this model; use a model with the required input support for visual tasks.
Requests above 32,000 input tokens may use a different price tier; the pricing section lists that tier separately.
Reported capabilities and production speed measurements describe access and observed traffic; task quality still needs an evaluation with your own examples.

Capability source · checked 2026-10-26

$1.755 estimated at displayed rates

1,000,000 ÷ 1,000,000 × $0.78 + 250,000 ÷ 1,000,000 × $3.90

The same billed quantities at the listed reference rates: $2.70.

Uses uncached input at the displayed standard rates, with no extra modalities or tools. Billed output includes any charged reasoning tokens. Actual token counts, price tiers, and marketplace prices can change the total.

Price quote effective Nov 1, 2026, 10:55 AM UTC.

Current model pricing and context compared with related models
Model	Current pricing	Context limit	Image input listed
qwen3-max-thinking (this model)Alibaba	$0.78 input / $3.90 output per 1M tokens	256,000 tokens	No
qwen3-next-80b-a3b-thinkingAlibaba	$0.0975 input / $0.585 output per 1M tokens	262,144 tokens	No
qwen3.7-maxAlibaba	$1.625 input / $4.875 output per 1M tokens	256,000 tokens	No
qwen3.8-max

qwen3-max-thinking

What you pay

Reliability and speed

Stop paying list price for tokens.

At a glance

Send a request

About qwen3-max-thinking

Workloads to evaluate

Limits and tradeoffs

Example: 1,000 text requests

Compare qwen3-max-thinking with related models

Parameters by endpoint

Pricing history

qwen3-max-thinking

When this model is cheapest

Popularity ranking

Product

Developers

Company

Legal