‹ BackHN Continuity

Thread

DeepSeek v4.1 Flash Is Now Our Best Hacking Model

177 points · 67 comments · talhof8

  1. habosa · · focus · HN ↗
    DeepSeek models have such good benchmark performance, amazing pricing, and the team over there seems to be widely considered impressive.

    I just haven't found them to be very good? I've had a ton more success with the GLM models (since 5.2 anyway). Maybe I'm just holding it wrong, DS models seem to get stuck in loops or tell me nonsense. GLM feels like budget Claude.

    1. tacomagick · · focus · HN ↗
      Depends on a lot of factors, AFAIK Claude Code (TUI Harness) does not work well with Deepseek, but Opencode is surprisingly good.
Open on Hacker News to reply ↗

Unofficial Hacker News client; not affiliated with Y Combinator.