From 8d035d8e4a88dfbf6948e9cd5776314110527faa Mon Sep 17 00:00:00 2001 From: Alvin Ryanputra Date: Tue, 25 Mar 2025 10:47:42 -0700 Subject: [PATCH] improvements to the readme --- README.md | 39 +++++++++++++++++++++++++-------------- 1 file changed, 25 insertions(+), 14 deletions(-) diff --git a/README.md b/README.md index fa79bc826..959d3c279 100644 --- a/README.md +++ b/README.md @@ -11,12 +11,18 @@

-[Codeflash](https://www.codeflash.ai) is a general purpose optimizer for Python that automatically improves the performance of your Python code while maintaining its correctness. -It uses advanced LLMs to generate multiple optimization ideas for your code, tests them to be correct and benchmarks them for performance. It then creates merge-ready pull requests containing the best optimization found, which you can review and merge. +[Codeflash](https://www.codeflash.ai) is a general purpose optimizer for Python that automatically improves the performance of your Python code while maintaining its correctness. -How to use Codeflash - -- Optimize an entire existing codebase by running `codeflash --all` -- Automate optimizing all __future__ code you will write by installing Codeflash as a GitHub action. +How Codeflash works: +1. LLMs generate multiple optimization candidates for your code +2. Codeflash tests the optimization candidates for correctness +3. Codeflash benchmarks the optimization candidates for performance + +Should the optimization be valid and faster than the original code, Codeflash will create a pull request with the optimized code. You can now review and merge the code to make your codebase faster! + +Ways to use Codeflash: +- Optimize an entire codebase by running `codeflash --all` +- Automatically optimize all __future__ code written by installing Codeflash as a GitHub action. Codeflash will try to optimize your new code before you merge it into the codebase. - Optimize a Python workflow end-to-end by tracing the workflow. Codeflash is used by top engineering teams at [Pydantic](https://github.com/pydantic/pydantic/pulls?q=is%3Apr+author%3Amisrasaurabh1+is%3Amerged), [Langflow](https://github.com/langflow-ai/langflow/issues?q=state%3Aclosed%20is%3Apr%20author%3Amisrasaurabh1), [Albumentations](https://github.com/albumentations-team/albumentations/issues?q=state%3Amerged%20is%3Apr%20author%3Akrrt7%20OR%20state%3Amerged%20is%3Apr%20author%3Aaseembits93%20) and many others to ship performant, expert level code. @@ -35,21 +41,26 @@ Add codeflash as a development time dependency if you are using package managers ## Quick Start -1. To configure Codeflash for a project, at the root directory of your project where the pyproject.toml file is located, run: +1. Run the following command at the root directory of your project where the pyproject.toml file is located ``` codeflash init ``` - - It will ask you a few questions about your project like the location of your code and tests - - Ask you to generate an [API Key](https://app.codeflash.ai/app/apikeys) to access Codeflash's LLMs - - Install a [GitHub app](https://github.com/apps/codeflash-ai/installations/select_target) to open Pull Requests on GitHub. - - Ask if you want to setup a GitHub actions which will optimize all your future code. - - The codeflash config is then saved in the pyproject.toml file. - -2. Optimize your entire codebase: + This will set up basic configurations for your project, eg: + - Input a Codeflash API key (for access to LLMs) + - Enable a [GitHub app](https://github.com/apps/codeflash-ai/installations/select_target) to open Pull Requests on the repo + - [Optional] Setup a GitHub actions which will optimize all your future code. + + + +2. Optimize a file: + + ``` + codeflash --file ``` +3. Optimize your entire codebase: (This will run for a while and open PRs as it finds optimizations) + ``` codeflash --all ``` - This can take a while to run for a large codebase, but it will keep opening PRs as it finds optimizations. ## Documentation For detailed installation and usage instructions, visit our documentation at [docs.codeflash.ai](https://docs.codeflash.ai)