I tried deepseek v4 through open code at the weekend. I'm a daily Claude/Claude code user.
I tried to build something simple and while it got the job done the thinking displayed did not fill me with confidence. It was pages and pages of "actually no", "hang on", "wait that makes no sense". It was like the model was having a breakdown.
Bear in mind open code was also new to me so I could be just seeing thinking where I usually don't
> "actually no", "hang on", "wait that makes no sense"
Claude does the same thing, claude code just hides the thinking now
I usually like the answers generated by those flows.
And before that they summarized it. But yeah, thinking was always like that (when it first started, it almost just seemed like a scheme to massively increase token use..)
> It tried to build something simple and while it got the job done the thinking displayed did not fill me with confidence. It was pages and pages of "actually no", "hang on", "wait that makes no sense". It was like the model was having a breakdown.
It has been probanly trained to assess its own "thoughts" regularly and outputs those for the assesment results. I wouldn't worry much about the reasoning text contents, and it's nice to have them in contrast to the closed model "summaries", so it's easier to see what's going on.
Before CC and Codex removed thinking/verbose and hid most of it, both do that .
Yeah people aren’t aware that we don’t see the actual traces anymore lol
You can just use it through Claude Code, so you get to keep the system prompt and tooling you are used to.
3rd party models are a drop-in replacement with `ANTHROPIC_BASE_URL` in Claude Code, something people seem to miss right now. And contrary to what Anthropic might like to have you think, you don't need Opus 4.7 to run the harness to get similar performance.
https://api-docs.deepseek.com/quick_start/agent_integrations...
I feel the reasoning might be tuned for hard questions and not agentic work. I feel it overthinks, good for a very hard question, not for small incremental agentic steps. In theory, disabling thinking and using really well formed instruction, forcing it to still emit a bunch of tokens each step prior to taking action, could help. Only one way to find out though.
Opus 4.6 and GPT 5.4 do the same thing through GH Copilot and Bedrock. I get plenty of "Actually the simplest solution is ..., wait no, actually I should do ..., the best fix is ..."
Eh, you're seeing raw thinking tokens. With Claude <x> 4, and I think GPT-5 series, you are no longer seeing real thinking tokens, but "summarized" tokens that are probably highly different to the raw thinking.
use hide_thinking in opencode to get the claude experience :p
I see similar things using GLM 5.1 in pi.
I had to turn off thinking traces because it was just giving me anxiety looking at it.
> Bear in mind open code was also new to me so I could be just seeing thinking where I usually don't
Well there's your problem.
Edit: I remember seeing similar things with ChatGPT or Codex, although I can't remember in which context.