Comparing our Rust-based indexing and querying pipeline to Langchain
(bosun.ai)
We frequently get asked why we’re building text (code) processing software in Rust and not in Python since our performance would be bottlenecked by the LLM inference anyway. In this article I’ll do some exploration on the performance of our Rust based indexing and querying pipeline, and how it compares to Python’s Langchain. In this specific scenario our tool is significantly faster, which we didn’t really expect, so I dove in and found out why.
We frequently get asked why we’re building text (code) processing software in Rust and not in Python since our performance would be bottlenecked by the LLM inference anyway. In this article I’ll do some exploration on the performance of our Rust based indexing and querying pipeline, and how it compares to Python’s Langchain. In this specific scenario our tool is significantly faster, which we didn’t really expect, so I dove in and found out why.