Alibaba: qwen-plus-latest API
The dynamic endpoint for the Qwen-Plus series by Alibaba Cloud, consistently pointing to the most capable and stable recent snapshot. Currently powered by the Qwen3 architecture, it strikes an optimal balance between performance, latency, and cost. It supports native "Dual-Mode": Thinking Mode for complex logical reasoning, high-level math, and long-horizon planning, and Non-Thinking Mode for near-instant responses. Supporting up to 1 million (1M) tokens, it excels at massive document analysis, repository-level coding, and intricate multi-turn agentic workflows.
- Context window: 1,000,000 tokens
- Max output: 32,768 tokens
- Input: text
- Output: text
- Reasoning: Supported
- Tool calling: Supported
- File input: Supported
- Released: 2025-07-29
- Knowledge cutoff: 2025-10-14
Frequently Asked Questions
What is the context window of qwen-plus-latest?
qwen-plus-latest supports a context window of up to 1,000,000 tokens.
Does qwen-plus-latest support function calling?
Yes. qwen-plus-latest supports tool / function calling.
Does qwen-plus-latest support reasoning?
Yes. qwen-plus-latest is a reasoning-capable model.
What is the knowledge cutoff of qwen-plus-latest?
The knowledge cutoff of qwen-plus-latest is 2025-10-14.