StackGPT

GitHub commit activity for top repositories in topic: "AI"

@unnamed1tw @maximehugodupre @cursor_ai Please address this. The lack of clarity is leading too many people to churn. I'm personally considering.

@StackGPT @maximehugodupre @cursor_ai Similar on other plans. The amount you pay gives you regular usage with additional bonus on top.

So... how much usage is included with @cursor_ai? I thought it was only $400...

@unnamed1tw @maximehugodupre @cursor_ai what about other plans ?

@maximehugodupre @cursor_ai It’s $400 plus bonus on us. The actual amount varies depending on model usage as well as Auto mode. We will show it better in next update.

@mervenoyann Good stuff!! What would you use for parsing huge tables (non-machine readable) e.g. 30x20 cells ?

after the OCR blog we dropped yesterday people still ask me for the "best" model under the post 😅 we wrote the blog because there's no single best OCR model 🤠 but there is a model best for your use case! so please read the blog 🙇🏻‍♀️ let us know what you think!

@openrouter Amazon Titan

New Stealth model live: "Cypher Alpha" 🥷 🧩 General-purpose model with built-in tool-calling support! 🎁 Completely free: start experimenting! ⚠️ Privacy Notice: prompts and completions are logged by the model creator and may be used for training

@cline @openrouter Amazon Titan

🚨 New stealth (free) model from @openrouter 🚨 Cyber Alpha features a 1M token context window and a maximum output of 10k tokens. The last stealth model from OpenRouter was GPT-4.1. Who do you think is behind this latest stealth model release?

@tom_doerr do you know if there's a similar project with a nice web UI instead of cli? mcp support would also be the cherry on the top

@czue had the same thought yesterday.. claude 3.5 wrote 500 lines of react state management spaghetti code while it could have just suggested to use redux or the likes

My biggest complaint about coding with LLMs—to the extent that I'm allowed to complain about magic technology that makes my life 100x easier—is that they should be better at recognizing when the question itself is dumb and telling me there's an alternative I don't know about.

@SebastianB929 wait till you have to use Google ads.. it's a nightmare

Seriously is there a single UX engineer at google? Like how bloated can you make a UI? Just have a look at the amount of icons in the Sidebar😅 And I tought Azure was bad 😂

@tom_doerr .. and vs @cursor_ai ?

Aider already saved me so much time today. There are a few advantages it has over OpenDevin: - Fast feedback since Aider only does a single round of editing - Low cost since you not only save tokens by having fewer iterations, but also produce way less context - Fast speed per iteration since way less context needs to be encoded - Better Models (Sonnet) for less money than running OpenDevin with gpt-4o-mini to get the work done - Voice mode - You can let it use it outside a sandbox since you run the code yourself

@tom_doerr not convinced yet! can u install it in a T3 full stack app and ask it to create a new page together with backend logic? :p

Aider pair programmer setup

@tom_doerr recently moved from @GitHubCopilot to @cursor_ai after a long time of ignoring it. let's see how long i can keep ignoring aider :)

Just found Aider, looks like it just came out a few weeks ago and already has 14k+ GitHub stars. It doesn't just edit just edit files but also commits the changes it did. The biggest advantage over tools like OpenDevin might be that Aider creates a map of your repo, should make it more practical for work on existing code.

Hey @elevenlabs! What's this page in your docs?

@tom_doerr can't wait to see if it gets better with L3-405B

@StackGPT I would expect that there is a prompt it could use once it keeps repeating to solve the issue instead of just terminating

OpenDevin with gpt-4o-mini surprisingly often gets stuck in a loop. Feels like this should be avoidable with more intelligent prompt engineering

@tom_doerr I attended part of their webinar but i still cannot get the idea.. does it have the capability of updating memories? because the sample code looked quite basic.. you query the database with a query, get a list of memories, then u update them by id.

Just reached over 10k GitHub stars

@Erwin_AI You missed Volkswagen's erWin which is German acronym for: Electronic Repair and Workshop INformation system :D

Okay so.. My name is Erwin AI. But for some reason, Erwin is now also a product, and they recently added AI to their product. And now that's the first thing you find in Google ffs How is "erwin" even a brand? It sounds weird as a brand name. But maybe that's just me..

Just demoed a new tool that lets you control financial charts using natural language commands. 🚀📈 #FinTech #Trading #AI #Charts

@nixcraft 40MB

How much hard drive storage did your first home computer have? 🤔💾

@FastAPI @denicmarko I've got you @FastAPI

@denicmarko 🥲🙋‍♂️

Backend Frameworks with the most stars on GitHub: 1. Django - 77.8K⭐️ 2. Laravel - 77.5K⭐️ 3. Spring Boot - 73.6K⭐️ 4. Flask - 67K⭐️ 5. Nest - 65.6K⭐️ 6. Express - 64.3K⭐️ 7. Ruby on Rails - 55.3K⭐️ 8. Meteor - 44.1K⭐️ 9. Koa - 35K⭐️ Which one is your favorite?

@tom_doerr i think it's the feeling of a community.. i don't get that on github.. also, on discord i sometimes get instant responses to my stupid questions that i wouldn't risk my GH reputation for 😎

Why do so many projects have Discord/Slack? What kind of discussions couldn't also happen in the GitHub Discussion section or on Reddit? Makes a project so much worse when solutions/advice is buried somewhere and you can't find it with a search