Theo video detected = opinion rejected Also I generally dislike thinking models for coding and prefer faster models, so if you have something easy gemini 2.0 is good
What's Theo?
Gemini 2.5 Pro vs. Claude 3.7 Sonnet: Coding Comparison
261–270 of 336 posts
Re: Gemini 2.5 Pro vs. Claude 3.7 Sonnet: Coding Comparison
#262Gemini 2.5 Pro set a wide SOTA on the aider polyglot coding leaderboard [0]. It scored 73%, well ahead of the previous 65% SOTA from Sonnet 3.7. I use LLMs to improve aider, which is >30k lines of python. So not a toy codebase, not greenfield. I used Gemini 2.5 Pro for the majority of the work on the latest aider release [1]. This is the first release in a very long time which wasn't predominantly written using Sonne…
As someone who just adopted Cursor (and MCP) 2-3 weeks ago, Aider seems like a different world.
The examples of "create a new simple video game" cause me to glaze over.
Do you have a screencast of how you use aider to develop aider? I'd love to see how a savvy expert uses these tools for real-world solutions.
Re: Gemini 2.5 Pro vs. Claude 3.7 Sonnet: Coding Comparison
#263I had huge success letting Gemini 2.5 oneshot whole codebases in a single text file format and then split it up with a script. It's putting in work for like 5 minutes and spits out a working codebase, I also asked it to show of a little bit and it almost one shotted a java cloud service to generate pdf invoices from API calls, (made some minor mistakes but after feeding them back it fixed them) I basically use two sc…
Anything that can fit in a single LLM output is not a "codebase" it's just a start. Far too many people with no experience in real software projects think their little 1800 line apps are representative of real software development.
Re: Gemini 2.5 Pro vs. Claude 3.7 Sonnet: Coding Comparison
#264Gemini 2.5 Pro set a wide SOTA on the aider polyglot coding leaderboard [0]. It scored 73%, well ahead of the previous 65% SOTA from Sonnet 3.7. I use LLMs to improve aider, which is >30k lines of python. So not a toy codebase, not greenfield. I used Gemini 2.5 Pro for the majority of the work on the latest aider release [1]. This is the first release in a very long time which wasn't predominantly written using Sonne…
As someone who just adopted Cursor (and MCP) 2-3 weeks ago, Aider seems like a different world. The examples of "create a new simple video game" cause me to glaze over. Do you have a screencast of how you use aider to develop aider? I'd love to see how a savvy expert uses these tools for real-world solutions.
I actually get asked for screencasts a lot, so I made recently made some [0].
The recording of adding support for 100+ new coding languages with tree-sitter [1] shows some pretty advanced usage. It includes using aider to script downloading a collection of files, and using ad-hoc bash scripts to have aider modify a collection of files.
[0] https://aider.chat/docs/recordings/
[1] https://aider.chat/docs/recordings/tree-sitter-language-pack...
Re: Gemini 2.5 Pro vs. Claude 3.7 Sonnet: Coding Comparison
#265Earlier quoted context omitted.
I suspect it probably won't work, although it's not necessarily because an LLM architecture could never perform this type of work, but rather because it works best when the training set contains inordinate sample data. I'm actually quite shocked at what they can do in TypeScript and JavaScript, but they're definitely a bit less "sharp" when it comes to stuff outside of that zone in my experience. The ridiculous amoun…
> the cynicism isn't out of fear or panic, its disappointment over and over and over Very much this. When you criticize LLM's marketing, people will say you're a ludite. I'd bet that no one actually likes to write code, as in typing into an editor. We know how to do it, and it's easy enough to enter in a flow state while doing it. But everyone is trying to write less code by themselves with the proliferation of reusa…
>I'd bet that no one actually likes to write code
And you'd be wrong. I, for one, enjoy the process of handcrafting the individual mechanisms of the systems I create.
Re: Gemini 2.5 Pro vs. Claude 3.7 Sonnet: Coding Comparison
#266Very nice comparison but constrained to greenfield.
Would love to see a similar article that uses LLMs to add a feature to Gimp, or Blender.
Re: Gemini 2.5 Pro vs. Claude 3.7 Sonnet: Coding Comparison
#267Earlier quoted context omitted.
I don't think the Copilot product is a flop - they're doing quite well selling it along with GitHub and Visual Studio (Code). The best part about it, coding-wise, is that you can choose between 7 different models.
I think he's talking about Microsoft Copilot 365, not the coding assistant. Makes one wonder how much they are offering to the owner of www.copilot.com and why on God's green earth they would abandon the very strong brand name "Office" and www.office.com
Had to lookup office.com myself to see it; their office package is literally called MS Copilot.
Re: Gemini 2.5 Pro vs. Claude 3.7 Sonnet: Coding Comparison
#268In the Rubic's cube example, to solve the cube gemini2.5 just uses the memorized scrambling sequence: // --- Solve Function --- function solveCube() { if (isAnimating || scrambleSequence.length === 0) return; // Reverse the scramble sequence const solveSequence = scrambleSequence .slice() .reverse() .map((move) => { if (move.endsWith("'")) return move.slice(0, 1); // U' -> U if (move.endsWith("2")) return move; // U2…
Thank you. This is the insidious thing about black box LLM coding.
Re: Gemini 2.5 Pro vs. Claude 3.7 Sonnet: Coding Comparison
#269Here is a real coding problem that I might be willing to make a cash-prize contest for. We'd need to nail down some rules. I'd be shocked if any LLM can do this: https://github.com/solvespace/solvespace/issues/1414 Make a GTK 4 version of Solvespace. We have a single C++ file for each platform - Windows, Mac, and Linux-GTK3. There is also a QT version on an unmerged branch for reference. The GTK3 file is under 2KLOC.…
I suspect it probably won't work, although it's not necessarily because an LLM architecture could never perform this type of work, but rather because it works best when the training set contains inordinate sample data. I'm actually quite shocked at what they can do in TypeScript and JavaScript, but they're definitely a bit less "sharp" when it comes to stuff outside of that zone in my experience. The ridiculous amoun…
> if that really were happening, we'd see it.
You're right, instead what we see is the emergence of "vibe coding", which I can best describe as a summoning ritual for technical debt and vulnerabilities.
Re: Gemini 2.5 Pro vs. Claude 3.7 Sonnet: Coding Comparison
#270Is there a less biased discussion? The OP link is a thinly veiled and biased advert for something called composio and really a biased and overly flowery view of Gemini 2.5 pro. Example: “Everyone’s talking about this model on Twitter (X) and YouTube. It’s trending everywhere, like seriously. The first model from Google to receive such fanfare. And it is #1 in the LMArena just like that. But what does this mean? It me…
Zvi Moshowitz's blog [0] is IME a pretty good place to keep track of the state of things, it's well-sourced and in-depth without being either too technical or too vibes-based. Generally every time a model is declared the new best you can count on him to have a detailed post examining the claim within a couple days.