Comparison of Claude Sonnet 3.5, GPT-4o, o1, and Gemini 1.5 Pro for coding

[bobismyuncle]

https://www.qodo.ai/blog/comparison-of-claude-sonnet-3-5-gpt-4o-o1-and-gemini-1-5-pro-for-coding/

[WorkerBee28474]

The article seems like cork-sniffing to me, so I'll just link a couple coding LLM leaderboards that have actual numbers:

https://aider.chat/docs/leaderboards/

https://livebench.ai/

[Terretta]

This piece has characteristics of being LLM written or at least LLM edited.

It also makes unusually phrased assertions: "A key advantage of GPT-4o is its ability to stay updated with recent programming practices, libraries, and frameworks."

[airstrike → Terretta]

It also has multiple entries for the words "crucial" and "essential" which LLMs love to insert.

[petercooper]

This sort of dross makes me long for a downvote button (even if only for items that make it to the front page to avoid abuse). It's not flag-worthy but it's not front page material. A listed disadvantage of Gemini 1.5 Pro is "higher computational overhead" as if I'm running it locally or something..

[rmbyrro]

The TL;DR says all the obvious things every developer who has awareness already knew for some time.