Codex was meant to reset today, but after seeing that Gemini 3.6 Flash came out and testing it out, they decided not to reset.😭
I think I've figured out the secret of GLM5.2's huge improvements.🧐
I'm going to sleep now.
If this post gets more than 1,000 likes tomorrow, I'll write an article to reveal the reason.
Deepswe's benchmark results are my own experience.
I've used all models,
GLM 5.2 ≈ Claude Opus 4.6–4.7.
Kimi 2.7 code more like inference optimization.
Looking forward to K3.
Doubao-seed 2.1 Pro around 37% ≈ Gemini 3.5 Flash.
code are quite weak, but visual are strong.
Show more
I received festival gifts from @Alibaba_Wan at @Ali_TongyiLab thank for the thoughtful presents.
A charging adapter, a portable neck pillow, and a handheld fan.
