Chat Models

Alibaba Qwen

Qwen3 Coder Plus

Best for Coding

Coding-focused Qwen model with 1M context and 65K output; tool availability and token prices depend on the deployment region.

CodingRepo reasoning

At a glance

Know the model before you prompt.

Alibaba Qwen API specifications
Context window
1,000,000 tokens
Maximum output
65,536 tokens
Inputs → output
Text → Text
Knowledge cutoff
Not verified

API model ID: qwen3-coder-plus

Capabilities & boundaries

What it supports. Where the limits are.

Tool support requires the appropriate API integration; a supported tool is not automatically active in every chat.

Supported API features and tools

  • Text-based code generation and repository reasoning
  • 1M context with 65,536 maximum output
  • Prefix completion and caching on the Singapore reference deployment
  • Provider-documented coding-agent workflows where compatible tools are configured

Before you choose

  • Singapore function calling is listed as unsupported; Beijing support does not establish support elsewhere.
  • Singapore structured outputs, web search, batch inference and fine-tuning are also unsupported in the reviewed table.
  • The highest input band has a much higher output rate. Keep large-repository requests scoped and measure actual token use.
  • 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.

Reasoning behavior

The reviewed Coder Plus overview does not establish a separate thinking switch, default effort or reasoning budget. Do not transfer controls from Qwen3.6/3.7 hybrid models.

  • The alias is documented as equivalent to qwen3-coder-plus-2025-07-22. Use the exact version when comparing regressions.
  • Maximum input is 997,952 tokens. Published context and output ceilings are not a promise that both maxima can be consumed simultaneously.
  • Validate generated code with the actual compiler, tests and review process. The model cannot prove execution from a proposed command alone.

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

Plan a repository refactor

Define a narrow boundary and invariants first.

Review the supplied module map and code to plan this refactor. List the invariants that must remain true, the smallest affected file set and a staged sequence of changes. Specify tests and rollback conditions for each stage. Do not rewrite unrelated modules or assume access to files I have not provided.

Workflow 02

Locate a regression from a diff

Trace evidence instead of guessing at causes.

Use this failing test, stack trace and diff to identify the most likely regression. Cite the exact changed behavior, separate hypotheses from confirmed facts and propose a minimal fix. Add a test that would fail before the fix and pass afterward; do not claim it ran unless actual results are supplied.

Workflow 03

Review tool integration assumptions

Check the actual endpoint before promising agent behavior.

Audit this coding-agent configuration against the provider capability table I provide. Identify model IDs, region, tool-calling assumptions and unsupported response options. Suggest a minimal smoke test for each uncertain behavior. Do not infer endpoint support from a model's marketing description.

Developer reference

Alibaba Qwen API pricing

These are provider API reference prices, not EZ Ai Assist subscription prices.

View EZ Ai Assist plans
Singapore / International API reference · USD per 1,000,000 tokens
Token typePrice
Input · input ≤32K$1.00
Output · input ≤32K$5.00
Input · 32K < input ≤128K$1.80
Output · 32K < input ≤128K$9.00
Input · 128K < input ≤256K$3.00
Output · 128K < input ≤256K$15.00
Input · 256K < input ≤1M$6.00
Output · 256K < input ≤1M$60.00
  • 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.
  • Across the four bands, implicit cached input is $0.20 / $0.36 / $0.60 / $1.20; explicit cache creation is $1.25 / $2.25 / $3.75 / $7.50 and reads are $0.10 / $0.18 / $0.30 / $0.60 per million tokens.

Common questions

A few things worth knowing.

Does coding-agent support guarantee tool calls?

No. The current table supports function calling in Beijing but lists it as unsupported in Singapore. Confirm the endpoint before configuring an agent.

Can I request structured output in Singapore?

The reviewed table says it is unsupported. A prompt asking for JSON does not create a schema-enforced API guarantee; validate returned text yourself.

Why list four pricing bands?

The Singapore rate changes at 32K, 128K and 256K input tokens. Above 256K, the listed output rate is $60 per million tokens; omitting that band would understate cost.

Is there a verified thinking toggle?

Not in the reviewed model-specific source. This guide does not copy a hybrid-thinking control from a different Qwen model.

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.

Same provider