兔子API 兔子API
中

model

Qwen

Direct answerQwen spans general, reasoning, coding, vision-language, and media generation; Model Studio publishes exact regional and protocol IDs, while third-party listings remain separate.

Updated · Reviewed

Current official portfolio

Model Studio currently lists Qwen 3.8 and 3.7 general lines plus dedicated coding, vision, image, audio, and video models. The same catalog also includes third parties such as DeepSeek, so selection must check the author rather than the host alone.

Protocol and region

A Qwen model may expose OpenAI-compatible, Anthropic-compatible, and DashScope-native endpoints, with base URLs varying by region and workspace. Compatibility labels do not make every parameter equivalent.

Choosing within the family

Capability-first or Max models fit complex engineering and long-running agents, Plus classes balance capability and cost, and Flash classes target high throughput; use dedicated code, vision, or media models where appropriate. Evaluate Chinese, tools, multimodality, and latency.

API use

This site uses adapted routes such as /v1/chat/completions with a Bearer token; direct Model Studio access uses the regional credential. Do not confuse a console display name with the exact model ID or another channel's namespace.

Limits and upgrades

Regions do not always launch simultaneously, and latest or undated IDs can move. Verify context, output, thinking, tools, vision, caching, residency, billing, and retirement.

Use cases

  • Chinese and multilingual
  • Coding and reasoning
  • Multimodality

API protocols

  • /v1/chat/completions

FAQ

Is the model name permanent?

No. Official aliases can move and enabled API IDs can change; pin a tested ID in production and verify availability separately in the model marketplace.

How should I choose among general, coding, vision, and media Qwen models?

Evaluate general and agent workloads across capability, latency, and throughput tiers, while preferring dedicated models for code, vision, image, audio, or video. Test on real Chinese, tool, and multimodal samples; parameters from one Qwen model do not automatically fit another.

Are all Model Studio listings Qwen, with identical protocols across regions?

No. Model Studio also hosts third-party models, so verify authorship. A Qwen model's region, workspace base URL, and OpenAI, Anthropic, or DashScope compatibility can differ, and direct-provider IDs are not interchangeable with gateway names.

  • Alibaba QwenFor direct Model Studio access, copy the API host for the target region and workspace and use a key from the same scope. New pay-as-you-go keys often start with sk-ws-, plan keys with sk-sp-, while this gateway uses its own token.
  • OpenAI-compatible API guideThis site's OpenAI-compatible routes use a Bearer token; Responses, Chat Completions, and media endpoints have distinct objects and streaming semantics and must match the target model.
  • Choosing and calling coding modelsSelect coding models by patch correctness, test pass rate, tool success, latency, and cost on real repositories, not marketplace copy or a single benchmark.

Official sources

  1. Qwen Models Official