3 ms·
I'm finding myself using Claude more and more, mostly because it's been more stable the past couple of weeks (for me, in comparison to ChatGPT), and gives simil
by martinky24 2y ago
I'm finding myself using Claude more and more, mostly because it's been more stable the past couple of weeks (for me, in comparison to ChatGPT), and gives similar quality results for my type of queries.
I'd highly recommend this to others.
- rubymamis 2y agoIt's also definitely better sometimes.
- 404mm 2y agoSo in other words, 60% of the time it works better every time? /s
- r2_pilot 2y agoShhh... Let's keep this to ourselves. I'm not eager to see Claude end up with the same fate
- ffhhj 2y agoI don't know why but Poe's Assistant is better at coding than the newer models like Gemini. What model is it based on?
- iknowstuff 2y agoIdk man I paid for the Claude’s best model and I’m unimpresssed. GPT4 and 4o (which seems dumber especially with translation, but more fun with images and everyday conversation) are much better experiences. Their app is so good too.
- 404mm 2y agoI found Claude responses slightly better than GPT4 but I kept getting annoyed by two issues: 1. You cannot edit your message. This seemingly silly feature is a very powerful tool on GPT. You can rephrase or adjust your question based on previous response and let it regenerate. 2. “I apologize but I don’t feel comfortable to discuss X or Y. The threshold is unreasonably high imo. These two together ended up being a deal breaker for me.
- ChrisArchitect 2y agoRelated: https://status.anthropic.com/incidents/gm6vlw4gc377 https://status.anthropic.com/incidents/gm6vlw4gc377 https://news.ycombinator.com/item?id=40575539 https://news.ycombinator.com/item?id=40575539