ParaLLeM is a library for orchestrating agentic LLM workflows using the Batch API.
We are scalable, traceable, concise, readable, and expressive.
We are developer-centered and lightweight.
We help you save 50% on all token costs.
| Sync/Batch | OpenAI | Anthropic | |
|---|---|---|---|
| Simple | ✅ | ✅ | ✅ |
| Structured Output | ✅ | ✅ | ✅ |
| Function Calls | ✅ | ✅ | ✅ |
| Web Search | ✅ | ✅ | ✅ |
| Image Input | ✅ | ✅ | ✅ |
| MCP | ✅ | - | - |
- We are built from the ground up around the Batch API.
- Switch between sync (sequential) and batch in just 1 line of code.
- Control flow is best represented with Python, not data structures.
- An agent is more than just one LLM.
Please refer to the documentation.