Please share your thoughts on it. I’m curious about the maximum number of lines/files it can process in one go. How capable is it?
I feel his request is inexpensive, but tool calls are a bit expensive
I can say it’s better than the normal 3.7 but the tool calls are too much.
I noticed that linter errors happen a lot when the context grows large but I thought 3.7 MAX was supposed to work better with large context. But I still see lots of linter errors.
My current usage below, you see 9 prompts but 75 tool calls.
| 75 premium tool calls * 5 cents per tool call | $3.75 | 
|---|---|
| 3 claude-3.7-sonnet-max requests * 5 cents per such request | $0.15 | 
| 6 claude-3.7-sonnet-thinking-max requests * 5 cents per such request | $0.30 | 
              
              
              2 Likes
            
            
          
              
              
              1 Like
            
            
          pretty cool example where I had a database migration issue in my flask app and 3.7-thinking-max ended up writing a python script to clean up the migrations, completely solved the issue, then cleaned up the temp script
thought it was pretty impressive
              
              
              1 Like
            
            
          Thank you
Did it fail in the same way with a .mdc run rules file in place?