Supported Clients¶
Adastra LLMGW supports multiple ways to interact with LLMs. Choose the client that best fits your use case and technology stack.
Available Clients¶
| Client | Use Case | Endpoint |
|---|---|---|
| Azure OpenAI | Azure-native deployments with full OpenAI feature support | /azure-open-ai |
| OpenAI | Standard OpenAI API compatibility, simple integrations | /openai |
| AWS Bedrock | Access to Claude, Titan, and other AWS Bedrock models | /aws-bedrock |
| Anthropic | Native Anthropic SDK access to Claude models on AWS Bedrock and Azure | /anthropic |
| Anthropic Claude in Office Add-in | Anthropic's Claude add-in for Word/Excel/PowerPoint pointed at LLMGW | /anthropic |
| Vertex AI | Access to Google Gemini and Model Garden models via OpenAI-compatible API | /openai |
| Langchain | Complex AI applications with advanced patterns and agents | Provider-specific |
| Azure Cohere | Azure Cohere reranker support | /azure-cohere |
Quick Comparison¶
- Azure OpenAI: Best for organizations using Azure infrastructure
- OpenAI: Best for straightforward integrations and broad compatibility
- AWS Bedrock: Best for accessing Claude models and AWS ecosystem tools
- Anthropic: Best for native Anthropic SDK usage with Claude on AWS Bedrock and Azure
- Anthropic Claude in Office Add-in: Best for end users in Word/Excel/PowerPoint
- Vertex AI: Best for accessing Google Gemini and self-deployed models in Google Model Garden
- Langchain: Best for building sophisticated AI workflows
- Azure Cohere: Cohere reranker models
Next Steps¶
Select a client guide above to get started, or check Response Headers to learn about monitoring and debugging your requests.