Finally, an openrouter for agent harnesses.
Devs just open-sourced a plug-and-play infrastructure layer that lets you run different agent harnesses through one interface. Including: → Codex → Claude Code → Hermes → DeepSeek Harness → System One, powered by JEV → 9+ more
A chinese developer built a free AI stock analyst.
54,900 people have already copied it. It texts you buy or sell before the market opens. And 5 out of 6 people who like it also install it. The grid at the top of the video is those copies. On GitHub you like a project with a
Would you look at that.
Deepseek-v4.1-flash via api is still freaking fast, i'm getting above 280t/s. It's still fast, close to 300t/s is crazy! No other model is as fast as this, while being this good! But deepseek_ai is not done tho. For some reason they are thinking this
Off-peak halves the price and a cache hit cuts input far more
DeepSeek V4 Flash input is $0.44 per million on a peak cache miss, $0.007 on an off-peak cache hit. The usage block on every response reports both. Read it. Free guide:
You have reached the end of the archive
All of deepseek