Inferable is a developer platform that makes it easy to build and deploy reliable, secure LLM-based applications.
🚀 2-Minute Setup: Get started in minutes with our managed platform - wrap your existing functions, REST APIs and GraphQL endpoints as tools.
🤖 Managed Agent Runtime: Managed ReAct-like agent runtime powered by your own functions.
⚡️ Durable Tool Calling: Our durable execution engine helps you agents recover from tool-calling failures, load balance across your compute, and caches results for fast re-runs.
🛡️ Zero Network Config: No inbound connections or networking config required - your compute runs securely behind your firewall, and polls Inferable for instructions.
💻 Your Infrastructure: All functions run on your Infrastructure - Keep sensitive data and compute on-premise while leveraging our agent runtime. Inferable can't see or access any runtime information or environment variables.
🔌 Multiple Language Support: Native SDKs for TypeScript, Go, .NET and more coming soon - integrate with your existing codebase in minutes.
Each SDK directory contains its own README with specific installation instructions, quick start guide, and usage examples.
For language-specific quick start guides, please refer to the README in each SDK's directory:
| Feature | Node.js | Go | .NET | React |
|---|---|---|---|---|
| Register Functions | ✅ | ✅ | ✅ | ❌ |
| Create Runs | ✅ | ✅ | ✅ | ✅ |
| Feature | Node.js | Go | .NET |
|---|---|---|---|
| Blob results | ✅ | ❌ | ❌ |
| Mask results | ✅ | ❌ | ❌ |
| Cached results | ✅ | ❌ | ❌ |
| Call Timeouts | ✅ | ❌ | ❌ |
| Call Retries | ✅ | ❌ | ❌ |
| Call Approval (Human in the loop) | ✅ | ❌ | ❌ |
| Auth / Run Context | ✅ | ❌ | ❌ |
For comprehensive documentation on using Inferable AI, please visit our official documentation.
We welcome contributions to all our SDKs. Please read our contributing guidelines before submitting any pull requests.
All SDKs in this repository are licensed under the MIT License. See the LICENSE file for details.