There is, in my personal opinion, a reason that reasoning is not front and center. Partially related to distillation.
tudelo
Deepseek has shifted to my general go to, this being one of the factors. Most likely claude is making up some small details to reinforce what it thinks is right or optimal, but generally still right.
Only recent quirk in longer chats with Deepseek, watching the the thought process and output; it sometimes becomes Chinese. Also not sure if this is to intentionally hide the thought process, or the actual information needs to be looked up and explained in Chinese.
Loveispain
In what form are you using Claude? In Claude Code in their desktop app, you can enable thinking via the transcript drop-down, and while it's not the exact thinking the model does as that makes it susceptible to distillation, it does work well enough for the user to understand what is being done by the model.
satvikpendem
Maybe cheat is a one of the factors of model?
cyrbuzz
yeah this feels frustrating with claude, a black box inside a black box. good thing they got competitors
setnone
> Opus 4.6 remains the best model because of this.
Huh? Are you not trying other open-weight models that stream thinking traces?
The new DeepSeek-V4-Flash-0731 should clobber Opus 4.6 in a lot of tasks. Kimi K3 and GLM 5.2 feel like they stand toe-to-toe with Opus 4.8 in my experience. This probably isn't the last stupid decision that Anthropic stands on, you might as well hedge your bet and put some money into another inference provider and see how it goes.
comments (6)
tudelo
Only recent quirk in longer chats with Deepseek, watching the the thought process and output; it sometimes becomes Chinese. Also not sure if this is to intentionally hide the thought process, or the actual information needs to be looked up and explained in Chinese.
Loveispain
satvikpendem
cyrbuzz
setnone
Huh? Are you not trying other open-weight models that stream thinking traces?
The new DeepSeek-V4-Flash-0731 should clobber Opus 4.6 in a lot of tasks. Kimi K3 and GLM 5.2 feel like they stand toe-to-toe with Opus 4.8 in my experience. This probably isn't the last stupid decision that Anthropic stands on, you might as well hedge your bet and put some money into another inference provider and see how it goes.
bigyabai