Compatible with OpenAI API and SDK, supporting all major AI models. Build more efficient AI applications through a unified interface, smart routing, and a global inference network.
Everything you need to integrate and scale AI inference.
Migrate instantly without modifying existing workflows. Fully compatible with the OpenAI API and mainstream SDKs, reducing integration cost.
Access all leading models through a single unified API. Switch between models by changing one parameter, no SDK changes needed.
Global inference node providing fast AI inference for users worldwide. Low-latency access to frontier AI models.
Dramatically reduce inference costs through unified scheduling and model optimization. Build production-grade AI on any budget.
Capability-based model selection. Switch between reasoning, speed, and specialized tasks.
Production-ready features designed for seamless integration and observability.
Pick the capabilities you need. Free to start, upgrade as you grow.
Which models are included in each plan
| Model | Free | Starter | Pro | Business | Ultimate | Enterprise |
|---|---|---|---|---|---|---|
| Dubhe Fast | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
| Dubhe Code | ❌ | ✅ | ✅ | ✅ | ✅ | ✅ |
| Dubhe Agent | ❌ | ✅ | ✅ | ✅ | ✅ | ✅ |
| Dubhe Plus | ❌ | ✅ | ✅ | ✅ | ✅ | ✅ |
| Dubhe Vision | ❌ | ❌ | ✅ | ✅ | ✅ | ✅ |
| Dubhe Reasoning | ❌ | ❌ | ❌ | ✅ | ✅ | ✅ |