OpenPilot, an open-source AI coding assistant for VS Code
If you have tried GitHub's Copilot Chat, then this will look familiar to you. The main difference, however, is you can choose which LLM to talk to. So far OpenPilot supports GPT 3.5/4 and Google PaLM 2. I'd like to add support for more models (Claude 2, etc) but have been stuck on waiting lists so I haven't tested them. OpenPilot will give context to your prompts by automatically including relevant files from your workspace. You don't need to have the file open or manually highlight the code in question. It does this using 2 techniques: 1. All the files in your workspace are transformed…
What it does
In the maker’s words, at launch
If you have tried GitHub's Copilot Chat, then this will look familiar to you. The main difference, however, is you can choose which LLM to talk to. So far OpenPilot supports GPT 3.5/4 and Google PaLM 2. I'd like to add support for more models (Claude 2, etc) but have been stuck on waiting lists so I haven't tested them. OpenPilot will give context to your prompts by automatically including relevant files from your workspace. You don't need to have the file open or manually highlight the code in question. It does this using 2 techniques: 1. All the files in your workspace are transformed into embeddings and stored in a DB (Chroma). When you enter a prompt, OpenPilot performs a similarity search to try to find the files you are talking about. 2. It uses a two-step process when communicating to the LLM. It first asks the LLM what files it needs to see in order to answer your prompt, passing along the file paths that matched. This lets the model confirm whether the files are really relevant, so we don't pollute the context with irrelevant info. It also gives it a chance to specify any files that didn't semantically match, which it might be able to infer exist based on your prompt and the structure of your project given the file paths it has seen. Finally, the prompt is resubmitted along with the contents of the files it asked for. Another feature of OpenPilot that Copilot Chat doesn't (yet) have is the ability to apply diffs. In Copilot Chat, you can insert code at your cursor position. I think OpenPilot's diffs have more potential because you can 1) both add and remove lines, 2) make changes in multiple places, and 3) edit a file you don't have open, or even multiple files. Unfortunately, GPT 3.5 is pretty bad at formatting diffs, in my experience. This is why I've put it behind a toggle. Some work needs to be done here. I would appreciate any feedback. If there is interest in this, I'd like to continue iterating on it. An obvious next step would be adding support for a locally-running LLM. Along with better strategies for finding relevant context, and better UX in general. Thanks!
Does the same job
all alternatives →
More ai this month
the category →
I trained a 125M-parameter transformer to autocomplete piano performances in real time (~108 notes/sec on an iPhone 15). The idea is basically GitHub Copilot or Tabnine, except instead of prompting it with code, you prompt it by playing a few notes on a MIDI piano. The model then continues what you played, entirely on-device. The app is free if anyone wants to try it. Happy to answer questions about the model, training, Core ML, or the many things that didn't work.
AI · 16d ago · simedw.com
Astute▲585Automate your B2B brand going viral, with new media creators
AI · 18d ago · company-app.joinastute.com
Hey HN, Henry from Cactus here! We previously released Cactus Needle, a 14MB agentic LLM for tool call, device use, and structured extraction for phones, wearables, smart homes, small robots and microcontrollers. We got really great feedback here, and have now incorporated the suggestions to release Needle 2. The whole model is a single 14MB binary that runs a full session in 28MB of RAM; 45m parameters at 2bit compression. Needle hits 500 tokens/sec decode speed on a Raspberry Pi 5, sits between 400-1,500 tokens/sec on VR devices like Meta Quest 3S and Apple Vision Pro, and ranges…
AI · 26d ago · cactuscompute.com
Launched alongside, August 2023
the whole month →
Lottielab▲871Create and ship lottie animations to sites and apps faster
Dev tools · 2023 · lottielab.com
- LC
Outlines is a Python library that focuses on text generation with large language models. Brandon and I are not LLM experts and started the project a few months ago because we wanted to understand better how the generation process works. Our original background is probabilistic, relational and symbolic programming. Recently we came up with a fast way to generate text that matches a regex (https://blog.normalcomputing.ai/posts/2023-07-27-regex-guide...). The basic idea is simple: regular expressions have an equivalent Deterministic-Finite Automaton (DFA) representation. We…
AI · 2023 · github.com