Text generation

Choose the right text generation model for AI agents, chatbots, and document processing.

Recommended models for coding tools

We recommend qwen3.7-plus for balanced performance and cost, with full tool calling and a 1M-token context window for large codebases. For the strongest reasoning, choose qwen3.7-max.

Migrate from closed-source models

Map your current GPT, Claude, or Gemini model to an equivalent Bailian model.

	Closed-source examples	Bailian recommendation
Highest capability	GPT-5.5, Claude Opus 4.7, Gemini 3.1 Pro	`qwen3.7-max`
Balanced	GPT-5.4, Claude Sonnet 4.6, Gemini 3 Pro	`qwen3.7-plus`, `deepseek-v4-pro`, `glm-5.2`
Lightweight & low-cost	GPT-5.4-mini, Claude Haiku 4.5, Gemini 3.1 Flash	`qwen3.6-flash`, `deepseek-v4-flash`, `MiniMax-M2.5`

Use cases

Start with qwen3.7-plus for chatbots, content generation, summarization, and document processing — it balances performance, cost, and built-in tools with a 1M-token context window. To cut costs, switch to qwen3.6-flash, which offers similar capabilities at a lower price. For the strongest reasoning, use qwen3.7-max (1M context, higher cost).

Office productivity (non-coding)

For non-coding office tasks such as document drafting, email composition, meeting note summarization, and data analytics, start with qwen3.7-plus — it balances performance and cost with a 1M-token context window, Function Calling support, and built-in tools. To reduce costs, try qwen3.6-flash, which delivers near-flagship performance at a lower price with the same context length. For the strongest reasoning capability, choose qwen3.7-max (higher cost). For long-document processing such as reviewing multiple contracts, use qwen-long (10M-token context window).

Note

TONGYI Lingma and Qoder are AI coding tools designed for software development. They are not intended for general office productivity tasks.

Context window

1 million tokens is roughly 750,000 English words, or 8-10 novels.

For long documents or large codebases: qwen3.7-plus / qwen3.6-flash (1 million tokens).
For standard tasks: 128k-256k tokens is typically sufficient.

For context window details, visit the Models page.

China (Beijing) | Singapore | US | Frankfurt

Thinking mode

Step-by-step reasoning for multi-step math, code debugging, architecture planning, and legal cross-referencing.

Enable with the enable_thinking parameter, or use reasoning.effort in the Responses API to control thinking depth. All Qwen3+ models support this feature, most of which operate in a hybrid mode togglable per request.

See Deep thinking.

Function calling and built-in tools

Let the model take actions such as querying weather, searching databases, or booking meetings.

Function calling (custom tools that the model calls): Supported by all general-purpose models.
Built-in tools (web search, code interpreter, web scraping) with no configuration required.

See Tool calling.

Structured output

Forces valid JSON output, useful for extracting structured data like names and addresses from text.

See Structured output.

Batch inference

Process large volumes at lower cost when latency is not critical.

See Batch inference.

Recommended models

Model ID	Context	Thinking mode	Function Calling	Built-in tools	Structured output
`qwen3.7-max` View snapshots `qwen3.7-max-preview` `qwen3.7-max-2026-06-08` `qwen3.7-max-2026-05-20` `qwen3.7-max-2026-05-17`	1M	Supported	Supported	Supported	Supported
`qwen3.7-plus` View snapshots `qwen3.7-plus-2026-05-26`	1M	Supported	Supported	Supported	Supported
`qwen3.6-flash` View snapshots `qwen3.6-flash-2026-04-16`	1M	Supported	Supported	Supported	Supported
`deepseek-v4-pro`	1M	Supported	Supported	Unsupported	Unsupported
`deepseek-v4-flash`	1M	Supported	Supported	Unsupported	Unsupported
`glm-5.2`	198k	Supported	Supported	Unsupported	Supported
`kimi-k2.6`	256k	Supported	Supported	Unsupported	Unsupported
`MiniMax-M3`	192k	Supported	Supported	Unsupported	Unsupported
`mimo-v2.5-pro`	1M	Supported	Supported	Unsupported	Supported

Legacy models

For new projects, use the Qwen3.6 or Qwen3.5 series. The following models are legacy and no longer recommended. Visit the Models page to view detailed model parameters, such as context window and billing.

Qwen3.6

Model ID	Context	Thinking mode	Function Calling	Built-in tools	Structured output
`qwen3.6-max-preview`	256k	Supported	Supported	Unsupported	Supported
`qwen3.6-plus` View snapshots `qwen3.6-plus-2026-04-02`	1M	Supported	Supported	Supported	Supported

Qwen3.5

Model ID	Context	Thinking mode	Function Calling	Built-in tools	Structured output
`qwen3.5-plus` View snapshots `qwen3.5-plus-2026-02-15`	1M	Supported	Supported	Supported	Supported
`qwen3.5-flash` View snapshots `qwen3.5-flash-2026-02-23`	1M	Supported	Supported	Supported	Supported
`qwen3.5-397b-a17b`	256k	Supported	Supported	Supported	Supported
`qwen3.5-122b-a10b`	256k	Supported	Supported	Supported	Supported
`qwen3.5-27b`	256k	Supported	Supported	Supported	Supported
`qwen3.5-35b-a3b`	256k	Supported	Supported	Supported	Supported

Qwen3

Model ID	Context	Thinking mode	Function Calling	Built-in tools	Structured output
`qwen3-max` View snapshots `qwen3-max-2026-01-23` `qwen3-max-preview` `qwen3-max-2025-09-23`	256k	Supported	Supported	Supported	Supported
`qwen3-235b-a22b`	256k	Supported	Supported	Unsupported	Supported
`qwen3-235b-a22b-thinking-2507`	256k	Supported	Supported	Unsupported	Unsupported
`qwen3-235b-a22b-instruct-2507`	256k	Unsupported	Supported	Unsupported	Supported
`qwen3-next-80b-a3b-thinking`	256k	Supported	Supported	Unsupported	Unsupported
`qwen3-next-80b-a3b-instruct`	256k	Unsupported	Supported	Unsupported	Supported
`qwen3-32b`	256k	Supported	Supported	Unsupported	Supported
`qwen3-30b-a3b`	256k	Supported	Supported	Unsupported	Supported
`qwen3-30b-a3b-thinking-2507`	256k	Supported	Supported	Unsupported	Unsupported
`qwen3-30b-a3b-instruct-2507`	256k	Unsupported	Supported	Unsupported	Supported
`qwen3-14b`	256k	Supported	Supported	Unsupported	Supported
`qwen3-8b`	256k	Supported	Supported	Unsupported	Supported
`qwen3-4b`	256k	Supported	Supported	Unsupported	Supported
`qwen3-1.7b`	256k	Supported	Supported	Unsupported	Supported
`qwen3-0.6b`	256k	Supported	Supported	Unsupported	Supported

Qwen3-Coder

Model ID	Context	Thinking mode	Function Calling	Built-in tools	Structured output
`qwen3-coder-plus` View snapshots `qwen3-coder-plus-2025-09-23` `qwen3-coder-plus-2025-07-22`	1M	Supported	Supported	Unsupported	Supported
`qwen3-coder-flash` View snapshots `qwen3-coder-flash-2025-07-28`	1M	Supported	Supported	Unsupported	Supported
`qwen3-coder-next`	256k	Supported	Supported	Unsupported	Supported
`qwen3-coder-480b-a35b-instruct`	256k	Unsupported	Supported	Unsupported	Supported
`qwen3-coder-30b-a3b-instruct`	256k	Unsupported	Supported	Unsupported	Supported

Qwen2.5 (open source)

Model ID	Context	Thinking mode	Function Calling	Built-in tools	Structured output
`qwen2.5-omni-7b`	1M	Unsupported	Unsupported	Unsupported	Unsupported
`qwen2.5-vl-72b-instruct`	1M	Unsupported	Unsupported	Unsupported	Unsupported
`qwen2.5-vl-32b-instruct`	1M	Unsupported	Unsupported	Unsupported	Unsupported
`qwen2.5-vl-7b-instruct`	1M	Unsupported	Unsupported	Unsupported	Unsupported
`qwen2.5-vl-3b-instruct`	1M	Unsupported	Unsupported	Unsupported	Unsupported
`qwen2.5-72b-instruct`	1M	Unsupported	Supported	Unsupported	Supported
`qwen2.5-32b-instruct`	1M	Unsupported	Supported	Unsupported	Supported
`qwen2.5-14b-instruct`	1M	Unsupported	Supported	Unsupported	Supported
`qwen2.5-14b-instruct-1m`	1M	Unsupported	Supported	Unsupported	Supported
`qwen2.5-7b-instruct`	1M	Unsupported	Supported	Unsupported	Supported
`qwen2.5-7b-instruct-1m`	1M	Unsupported	Supported	Unsupported	Supported

Translation

Model ID	Context	Thinking mode	Function Calling	Built-in tools	Structured output
`qwen-mt-plus`	16k	Unsupported	Unsupported	Unsupported	Unsupported
`qwen-mt-turbo`	16k	Unsupported	Unsupported	Unsupported	Unsupported
`qwen-mt-flash`	16k	Unsupported	Unsupported	Unsupported	Unsupported
`qwen-mt-lite`	16k	Unsupported	Unsupported	Unsupported	Unsupported

Qwen-Long

Model ID	Context	Thinking mode	Function Calling	Built-in tools	Structured output
`qwen-long` View snapshots `qwen-long-2025-01-25`	10M	Unsupported	Unsupported	Unsupported	Supported
`qwen-long-latest`	10M	Unsupported	Unsupported	Unsupported	Supported

Role-playing

Model ID	Context	Thinking mode	Function Calling	Built-in tools	Structured output
`qwen-plus-character`	32k	Unsupported	Unsupported	Unsupported	Unsupported
`qwen-plus-character-ja`	32k	Unsupported	Unsupported	Unsupported	Unsupported
`qwen-flash-character`	8k	Unsupported	Unsupported	Unsupported	Unsupported

Legacy Qwen

Model ID	Context	Thinking mode	Function Calling	Built-in tools	Structured output
`qwen-plus` and its snapshots	1M	Supported	Supported	Supported	Supported
`qwen-max` and its snapshots	128k	Unsupported	Supported	Supported	Supported
`qwen-flash` and its snapshots	1M	Supported	Supported	Supported	Supported
`qwen-turbo` and its snapshots	1M	Supported	Supported	Supported	Supported
`qwq-plus`	128k	Supported	Supported	Unsupported	Unsupported
`qvq-max` and its snapshots	128k	Supported	Unsupported	Unsupported	Unsupported
`qwen-omni-turbo` and its snapshots	32k	Unsupported	Unsupported	Unsupported	Unsupported

Third-party models

Model ID	Context	Thinking mode	Function Calling	Built-in tools	Structured output
`glm-5.1`	198k	Supported	Supported	Unsupported	Supported
`glm-5`	198k	Supported	Supported	Unsupported	Supported
`glm-4.7`	198k	Supported	Supported	Unsupported	Supported
`glm-4.5`	198k	Supported	Supported	Unsupported	Supported
`glm-4.5-air`	198k	Supported	Supported	Unsupported	Supported
`MiniMax-M2.7`	200k	Supported	Supported	Unsupported	Unsupported
`MiniMax-M2.5`	200k	Supported	Supported	Unsupported	Unsupported
`MiniMax-M2.1`	200k	Supported	Supported	Unsupported	Unsupported
`kimi-k2.5`	256k	Supported	Supported	Unsupported	Unsupported
`kimi-k2-thinking`	256k	Supported	Supported	Unsupported	Supported
`Moonshot-Kimi-K2-Instruct`	256k	Unsupported	Supported	Unsupported	Unsupported
`deepseek-v3.2`	128k	Supported	Supported	Unsupported	Unsupported
`deepseek-v3.2-exp`	128k	Supported	Supported	Unsupported	Unsupported
`deepseek-v3.1`	128k	Supported	Supported	Unsupported	Unsupported
`deepseek-v3`	128k	Unsupported	Supported	Unsupported	Unsupported
`deepseek-r1`	128k	Supported	Supported	Unsupported	Unsupported
`deepseek-r1-0528`	128k	Supported	Supported	Unsupported	Unsupported
`deepseek-r1-distill-llama-70b`	128k	Supported	Unsupported	Unsupported	Unsupported
`deepseek-r1-distill-qwen-32b`	128k	Supported	Unsupported	Unsupported	Unsupported
`deepseek-r1-distill-qwen-14b`	128k	Supported	Unsupported	Unsupported	Unsupported
`deepseek-r1-distill-qwen-7b`	128k	Supported	Unsupported	Unsupported	Unsupported
`deepseek-r1-distill-qwen-1.5b`	128k	Supported	Unsupported	Unsupported	Unsupported
`deepseek-r1-distill-llama-8b`	128k	Supported	Unsupported	Unsupported	Unsupported