GoKawiilAgentic LLMs like Opus 4.5 shown to auto-optimize Rust code 2x-20x faster
A developer tested whether repeatedly asking modern agentic LLMs to improve code could yield genuine performance gains, this time using Rust rather than Python. After months of experimentation following the release of Opus 4.5, the author found that these models can produce Rust code significantly faster than current state-of-the-art implementations, provided they are given proper constraints and guardrails.
GoKawiil's interpretation of the reporting above, not reported fact.
This suggests that as coding-focused LLMs improve, they may become useful not just for writing functional code but for systematically optimizing performance-critical software, a task traditionally requiring deep domain expertise. Because the author published prompts and benchmarks rather than vague claims, the results offer a reproducible signal of how far automated code optimization has progressed.
- Modern agentic LLMs like Opus 4.5 can meaningfully speed up Rust code when properly guided.
- Speedups ranged from 2x to 20x depending on the coding domain tested.
- The author shared actual prompts and benchmark data to make the claims verifiable, unlike earlier speculative posts.
Source: minimaxir.com, 2026-09-22
Published there as: “Writing Rust code that's fast by asking agents to make the code faster”
Read the original report → The summary and analysis above are GoKawiil's own, written from reporting by the source above. Facts and quotes belong to the original publisher.