[image] R1+Sonnet set SOTA on aider’s polyglot benchmark R1+Sonnet has set a new SOTA on the aider polyglot benchmark. At 14X less cost compared to o1. Greetings, I thought I’d bring this up: Apparently the team behind Aider has found out that combining R1’s…

R1+Sonnet combination implementation?

normalnormie January 28, 2025, 1:26am 8

Temporarily, using a manual approach could work(lets test it): Plan/Act instruction + changing model to Claude when sending Act

2 Likes

Topic		Replies	Views
R1 model is amazing Discussion	34	8513	February 12, 2025
Can i use cursor as in DeepClaude? they claim DS R1+C3.5 will already beat 3.7+thinking. they essentially same Discussion	1	65	February 28, 2025
O3-mini supported the second it's out. Why isn't Deepseek-R1 supported in Composer's Agent mode? Discussion	4	587	February 4, 2025
Plan Act Mode in Composer Agent Feature Requests	7	645	February 28, 2025
Deepseek R1 Agent Feature Requests	29	7090	March 9, 2025