batchlane

batchlane is a Python library that allows users to submit asynchronous batch jobs to multiple large language model (LLM) providers' discount batch lanes through a single unified interface. It supports providers such as Anthropic, Gemini AI Studio, OpenAI, Groq, Mistral, Fireworks, Together, and DeepInfra. The tool handles chunking, polling, merging results, and allows resumable job submissions with checkpointing support. It is designed to optimize cost and efficiency when batching LLM inference requests.