Tool support requires the appropriate API integration; a supported tool is not automatically active in every chat.
Supported API features and tools
Native text, image and video understanding
1M context and a 65,536-token output ceiling
Hybrid thinking enabled by default
Singapore function calling, structured outputs, search, prefix completion and caching
Before you choose
Singapore batch inference and fine-tuning are unsupported in the reviewed model table.
Input limits differ by mode: 991,808 without thinking and 983,616 with thinking.
No universal tokens-per-second guarantee or native media generation is established here.
The reviewed sources do not establish a knowledge cutoff. Supply current evidence for time-sensitive questions; a large context window is not live web access.
Extended thinking
Thinking enabled · defaultThinking disabled
This model supports thinking and non-thinking requests. Alibaba Cloud's enable_thinking switch controls the mode; in the OpenAI Python SDK it belongs in extra_body. The provider's thinking_budget controls reasoning length. These are integration settings, not features a prompt can activate in every app. Thinking is enabled by default for this version. Use a bounded budget and evaluate whether extra reasoning improves the task.
The thinking guide lists qwen3.6-flash and the qwen3.6-flash-2026-04-16 snapshot. Confirm the actual serving version when testing.
The published chain-of-thought ceiling is 131,072 tokens, separate from maximum answer output. Do not copy the Plus model's thinking ceiling.
Put it to work
Start with a more useful prompt.
Original examples from EZ Ai Assist. Adapt these to your task and the features available in your workspace.
Workflow 01
Classify visual issues
Use a compact schema for repeated image review.
Classify these supplied product screenshots using my issue categories. Return screenshot ID, category, visible evidence and whether manual review is needed. Keep uncertain cases separate and do not infer interactive failures from a still image. Use exactly the requested fields and no invented categories.
Workflow 02
Extract a short video timeline
Locate visible events without adding unstated context.
Create a concise timeline from this short video using the supplied event definitions. Include timestamps, observed actions and confidence notes. Mark events that cannot be verified from the clip. Do not identify people or claim intent; focus on visible and task-relevant evidence.
Workflow 03
Review a repetitive code transformation
Catch edge cases before scaling a workflow.
Review these before-and-after code pairs against the transformation rules. Report rule violations with line references, explain edge cases and propose minimal corrections. Design a small regression set before applying the transformation more broadly. Do not claim a repository-wide change has been performed.
Developer reference
Alibaba Qwen API pricing
These are provider API reference prices, not EZ Ai Assist subscription prices.
Singapore / International API reference · USD per 1,000,000 tokens
Token type
≤ 256Kinput tokens
> 256Kinput tokens
Input
$0.25
$1.00
Output
$1.50
$4.00
Explicit cache creation
$0.3125
$1.25
Explicit cache read
$0.025
$0.10
Rates describe the Singapore / International deployment, in USD per million tokens, excluding promotions. Other regions, snapshots and hosting services can have different rates and supported features.
Input-length bands refer to the provider's published prompt-size tiers, not the length of the answer. Check the selected region, resolved model version, cache policy and any additional service charges before estimating cost.
Common questions
A few things worth knowing.
Is Flash text-only?
No. This specific Qwen3.6-Flash model table lists image, text and video inputs with text output.
Is it always cheaper than Plus?
The displayed Singapore token rates are lower for these published bands, but total cost also depends on token use, retries, caching and task quality. Compare equivalent workloads.
Does it support the same thinking budget as Plus?
The table lists a different chain-of-thought ceiling: 131,072 for Flash versus 81,920 for Plus. Both list 65,536 maximum answer output.
Can I use batch inference in Singapore?
The reviewed Singapore capability table says no. Do not assume a Beijing feature or another provider's feature is enabled in this region.
Does this page guarantee access in EZ Ai Assist?
No. A catalog entry and provider capability reference do not establish current app access, enabled tools or available settings. Check the model selector and controls in your workspace.
Are these EZ Ai Assist subscription prices?
No. Provider API rates are developer references. Your subscription and any limits shown in the app are separate; use the website pricing page for plans.
How should I verify an answer?
Provide the relevant evidence and acceptance criteria, ask for source references or reproducible checks, and review the result. Distinguish a proposed tool action from a tool action actually executed.
Check the source
Official documentation
Specifications and API prices checked on . Example prompts and workflow advice are editorial guidance from EZ Ai Assist.