[HN Gopher] Show HN: CodeGPT.nvim - ChatGPT plugin for Neovim ___________________________________________________________________ Show HN: CodeGPT.nvim - ChatGPT plugin for Neovim Author : dpayne Score : 130 points Date : 2023-03-08 17:15 UTC (5 hours ago) (HTM) web link (github.com) (TXT) w3m dump (github.com) | madelyn wrote: | I just tried this out and it's basically exactly what I want! No | frills, basic feature set. Was going to make something myself but | now I can be even lazier than normal. | klinquist wrote: | This is a ChatGPT plugin for Neovim, not a Neovim plugin for | ChatGPT. | dang wrote: | Ok, we've swapped the order in the title above. | ilaksh wrote: | It defaults to temp 0.6. I thought that 0 was better for | following instructions? At least it was for text-davinci-003. Is | chatgpt-turbo different in this regard? | | Also if you have installed any GPT command line program like | askleo and don't need to give it context you can just run :r | !askleo description of function and language here | | Although askleo uses text-davinci-003 so I would use one set up | for ChatGPT instead. Or if you are not too lazy this extension | looks great. | | What would be really nice is something that automatically | selected context and then streamed in the code. Also it should | come with a short command or default for that. | qudat wrote: | Very cool! I added it to https://neovimcraft.com/?search=codegpt | renewiltord wrote: | Sick dude. Thank you! I love copilot.vim and use it quite | extensively. Will try this. I only ever use the ChatGPT service | through their website. Do you get an API key by default or is | that a separate service? | behnamoh wrote: | Can we have something like the OP's method in VSCode, WITHOUT | using CoPilot? | dpayne wrote: | I still use copilot as my main completion tool. I found ChatGPT | to be too slow and expensive to use for tab complete level of | usage, but it works really well for larger things (e.g. | refactor X to do Y, or write unit tests). | | You can get your API key here | https://platform.openai.com/account/api-keys. | renewiltord wrote: | Thank you! | jarenmf wrote: | Does the API offer access to the same model as the one used in | ChatGPT web app? If so it might be way cheaper to use the API | than ChatGPT plus? | capableweb wrote: | > Does the API offer access to the same model as the one used | in ChatGPT web app? | | Chat API is this: https://platform.openai.com/docs/api- | reference/chat/create | | URL that the plugin is using for the API call is here: https: | //github.com/dpayne/CodeGPT.nvim/blob/1ff70459b170ddaf... | | TLDR: yes | | > If so it might be way cheaper to use the API than ChatGPT | plus? | | I haven't used ChatGPT Plus, just the normal one. But at | least the demonstration GIF makes it seem kind of slow. One | selling point of the Plus service is supposedly that you get | faster responses. | jarenmf wrote: | I just tested the API, it seems to have the same speed as | ChatGPT plus. I quickly tested some questions and it seems | that OpenAI web app has a bit more relevant answers, I | wonder whether it's related to the temperature and other | parameters. | freedomben wrote: | It's not the same. I don't know the technical differences, | but in my tests it often returns wildly different results. | anecdotally the API has also hallucinated modules that don't | exist (but that I wish did!) whereas the chat gpt proper has | not done that to me. | QuadrupleA wrote: | I suspect it comes down to the "system" prompt the API lets | you set, which is hidden & pre-set on ChatGPT proper. | _emacsomancer_ wrote: | Also in Emacs: | | https://github.com/joshcho/ChatGPT.el | | https://youtu.be/SL-nNOjqoxg | robertkrahn01 wrote: | Another Emacs integration, using org-mode for ChatGPT and DALL-E: | https://github.com/rksm/org-ai | orsenthil wrote: | So, these utilize OpenAI API key. Do people run into limits have | to go for paid version while utilizing apps based on openai API? | aunch wrote: | this is really cool and well done. i will start using it in | conjunction with Codeium's plugin (free Copilot alternative): | https://github.com/Exafunction/codeium.vim | | for transparency: I'm from the Codeium team, and we are big fans | of getting this AI gen tech to all developers on all IDEs for | free - we've also open sourced an emacs plugin: | https://github.com/Exafunction/codeium.el | z3t4 wrote: | HN encurage you to mention your own products when relevant, so | you don't have to wrap it in, just put it bluntly. | capableweb wrote: | How are you supposed to use two plugins providing the same | functionality? Are you gonna compare them and then uninstall | one of them? Do they differ? | | As far as I can tell, it seems like you just wanted to plug | your own plugin, but I'm happy if I'm wrong. I just to | understand how you can use two plugins that provide the same | functionality and why you would do that. | aunch wrote: | they're complementary! codeium works like copilot, constantly | autocompleting what you want to type next, which helps a lot | when you already have a decent idea of what you need to get | done and need to get it done faster. chatgpt (and chat | interfaces in general) help you get some ideas on what to do | when you are doing more open-ended exploration, and is | explicitly invoked. as developers, we all code in these | different "modes" and different tools can be built for each! | kronion wrote: | See also ChatGPT.nvim, which can be used both for editing code | and for standard chat: https://github.com/jackMort/ChatGPT.nvim | | The Telescope-esque UI is nicely done as well. | kilgnad wrote: | It's like we're LLMs ourselves... generating different flavors of | UIs to interface with chatGPT. That's what I think of when I see | the proliferation of chatGPT UIs that hit the front page. | | Inevitably, human behavior that's similar to an LLM is ripe for | replacement by an LLM due to the abundance of behavioral data. | nico wrote: | > It's like we're LLMs ourselves... | | Absolutely. That's a very exciting thing. These tools are | forcing us to confront who we are. | | Interestingly, LLMs (so far) are similarly limited, they can | only ingest/digest so much data at a time, so we are trying to | figure ways around it. | | Just like humans, soon there will be lots of specialized LLMs, | ones that do only certain types of law, or only one programming | language, or even just one codebase, etc. | | Fun times coming. | QuadrupleA wrote: | Ha - this was my weekend project, was pretty sure something | similar would pop up in a few days. It's been very handy to ask | questions, summarize, proofread, give jailbroken wisecracks, etc. | right within VIM where I can slice and dice the text at will. | | Not sure if this supports the API's "streaming mode" but it's | nice for seeing the output typing in real time. Requires spawning | a separate job though. ___________________________________________________________________ (page generated 2023-03-08 23:00 UTC)