Platforms
Hashbrown uses the adapter pattern for supporting multiple platforms.
Official Adapters
| Platform | Adapter Package |
|---|---|
| OpenAI | @hashbrownai/openai |
| Microsoft Azure | @hashbrownai/azure |
| Anthropic Claude | @hashbrownai/anthropic |
| Amazon Bedrock | @hashbrownai/bedrock |
| Google Gemini | @hashbrownai/google |
| Ollama | @hashbrownai/ollama |
Custom Adapters
Can't find your preferred AI provider? Create a custom adapter for any LLM that supports streaming chat completions.
Platform Capabilities
| Platform | Text | Streaming | Tools | Structured Output |
|---|---|---|---|---|
| OpenAI | ✅ | ✅ | ✅ | ✅ |
| Microsoft Azure | ✅ | ✅ | ✅ | ✅ |
| Anthropic Claude | ✅ | ✅ | ✅ | ✅ |
| Amazon Bedrock | ✅ | ✅ | ✅ | ✅ |
| Google Gemini | ✅ | ✅ | ✅ | ✅ |
| Ollama | ✅ | ✅ | ✅ | ✅ |
Platform Limitations
| Platform | Limitations |
|---|---|
| OpenAI | None |
| Microsoft Azure | None |
| Anthropic Claude | Requires @anthropic-ai/sdk peer dependency |
| Amazon Bedrock | Native structured output support varies by model |
| Google Gemini | Native structured output support varies by model |
| Ollama | Limited model support |
Where is X platform?
Need a platform that is not listed here? Open an issue on GitHub and tell us what you are building. If your team wants a finished, headful agent UI with enterprise support, take a look at threadplane.