Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
345f506fdc718
searching Neon…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
5 ms
·
1.
▲
by
345f506fdc718
2mo ago
Maybe it got a lot less and I just got used to it. True. Still feels too much for me. Breaks my workflow for no reason. Too much overhead for me, if I can't trust the output
2.
▲
by
345f506fdc718
2mo ago
> 15 months ago, the top model on the Artificial Analysis index was GPT-o3. It scores 30 on the Artificial Analysis index. There must be something really of with those benchmarks. Yes, hallucinations gotten better, but I don't see t