hellno the optimist
@hellno.eth
DeepSeek R1 after one day of coding for >4hrs → much better than Claude and much cheaper. R1 has a small delay, where before results would stream immediately. The pause is not annoying so far, but tricks me into believing the results are better than models who don't 'reason'. 10/10 would LLM again
3 replies
0 recast
5 reactions
Samuel ツ
@samuellhuber.eth
how do you use it? cursor?
1 reply
0 recast
0 reaction
hellno the optimist
@hellno.eth
https://aider.chat/
1 reply
0 recast
1 reaction
