3 ms·

> My own experience w/ 4.6 and 4.7 are that I don't firmly grasp any capabilities improvements over my memory of 4.5, but it's all so fuzzy that it's truly diff

> My own experience w/ 4.6 and 4.7 are that I don't firmly grasp any capabilities improvements over my memory of 4.5, but it's all so fuzzy that it's truly difficult to tell. I've actually intentionally switched back to 4.5. I hated 4.7 so much that I decided to jump back all the way to 4.5. Now that I've been using 4.5 for a few weeks, I find it significantly more reliable but a bit more forgetful than 4.6/4.7. I'm okay with that because it's really easy to identify this forgetfulness and nudge it. I found 4.7's adaptive thinking to be extremely unreliable. It seems to overcorrect on the current message without considering the difficult of the overall problem. I wonder if 4.8 will improve on that.

dwaltrip 5mo ago
If you are using Claude code, just set effort to xhigh. This one change will probably solve 80% of the problems you have noticed.
- orwin 5mo ago
  This. XHigh and the 'plan' mode for complex tasks is absolutely a must have. Still, the context window is sometimes too small for my usage.
  - jayGlow 5mo ago
    agent teams can help with that, the main agent acts as an orchestrator and spawns sub agents to do the actual tasks it generally keeps the main context from overflowing.
- whatevaa 5mo ago
  Isn't xhigh on opus 4.7 very expensive on tokens?
  - dwaltrip 5mo ago
    I’ve never ran into the limits on the $100 plan, and rarely even get close. I normally have only one session going at once though.
    - joshstrange 5mo ago
      Same here and while I have multiple sessions going from time to time, my day isn't spent primarily developing software directly anymore (due to role, nothing about LLMs). I only ever hit the $100/mo limits 1-2 times ever and it was always <1hr before reset (once it was <5min, the other was like ~45min). I'm even considering going back down to $20 and using extra usage for the times I need to "burst".
  - sumedh 5mo ago
    Yes but Anthropic made a deal with SpaceX and increase usage limits by 50%, so you might not hit your limits.
michaelsalim 5mo ago
Same here. Went back to 4.5 and was happy I did it. The only frustration was that I can tell the model has declined compared to the first few weeks it was released. I also recently moved to 4.6 since I started hitting the context limit too often with my current project.
- luxuryballs 5mo ago
  /model claude-opus-4-6[1m] allows you to specify you want the 1 million context 4.6

dwaltrip 5mo ago

If you are using Claude code, just set effort to xhigh. This one change will probably solve 80% of the problems you have noticed.

orwin 5mo ago
This. XHigh and the 'plan' mode for complex tasks is absolutely a must have. Still, the context window is sometimes too small for my usage.
- jayGlow 5mo ago
  agent teams can help with that, the main agent acts as an orchestrator and spawns sub agents to do the actual tasks it generally keeps the main context from overflowing.
whatevaa 5mo ago
Isn't xhigh on opus 4.7 very expensive on tokens?
- dwaltrip 5mo ago
  I’ve never ran into the limits on the $100 plan, and rarely even get close. I normally have only one session going at once though.
  - joshstrange 5mo ago
    Same here and while I have multiple sessions going from time to time, my day isn't spent primarily developing software directly anymore (due to role, nothing about LLMs). I only ever hit the $100/mo limits 1-2 times ever and it was always <1hr before reset (once it was <5min, the other was like ~45min). I'm even considering going back down to $20 and using extra usage for the times I need to "burst".
- sumedh 5mo ago
  Yes but Anthropic made a deal with SpaceX and increase usage limits by 50%, so you might not hit your limits.

michaelsalim 5mo ago

Same here. Went back to 4.5 and was happy I did it. The only frustration was that I can tell the model has declined compared to the first few weeks it was released. I also recently moved to 4.6 since I started hitting the context limit too often with my current project.

luxuryballs 5mo ago
/model claude-opus-4-6[1m] allows you to specify you want the 1 million context 4.6