Qwen is Alibaba's large language model family. APIs are exposed via Alibaba Cloud Model Studio (DashScope) with per-token pricing across multiple regions. The qwen.ai consumer site fronts the chat product.
URL: Visit APIs.json URL
Run: Capabilities Using Naftiko
- x-type: company
- AI, LLM, Inference, Open Source, Alibaba, Multimodal
- DashScope (Alibaba Cloud Model Studio) API — Chat, vision, embeddings, audio (TTS/ASR), image generation, video generation, function calling. DashScope-native + OpenAI-compatible endpoints. Regional base URLs across Singapore (
dashscope-intl), US-Virginia, HK, Frankfurt, Beijing. Docs · Pricing · GitHub
Qwen3-Max, Qwen3.5-Plus, Qwen3.5-Flash, Qwen3-VL-Plus / Flash, Qwen3-Coder, Qwen-Plus / Flash (China), Qwen-VL-OCR. Open-weights: Qwen3-32B/14B/7B, Qwen2.5-72B/32B/Coder.
- Plans — PAYG tiered by context window. Qwen3-Max $1.2/M input @ 32K; Qwen3.5-Flash $0.10/M @ 128K. Free self-hosting for open weights.
- RateLimits — Per-model RPM/TPM and concurrent-task quotas in workspace console.
- FinOps — FOCUS-aligned, billed via Alibaba Cloud invoice.
- Created: 2026-05-08
- Modified: 2026-05-08
- Public OpenAPI spec was not discovered for DashScope at the time of profiling; the OpenAI-compatibility surface follows the standard OpenAI spec shape.
FN: Kin Lane
Email: kin@apievangelist.com