As soon as the price of DeepSeek increased, Liang Wenfeng was downgraded from "Liang Sheng" to "Xiao Liang"!
After the launch of V4 Pro 0813, the community continued to dispute the actual test experience and version stability; the API was changed to peak-valley billing, and the price/performance ratio that was once "selected with eyes closed" began to be recalculated; Harness's plug-in and Agent base are very ambitious, but many people think it is too geeky. DeepSeek…
Last update on my tui for deepseek harness before i sleep.
Cleaned up tool calls and added settings to choose the amount of details shown.
I tested GLM 5.3 on the same tasks I used for the other models yesterday.
For complex animated 3D design, it is still far behind Kimi K3 and Opus 5. The lack of vision is still a real limitation here. It is also very slow, even slower than GLM 5.2. The same test took Opus,
Gemini 3.7 Flash vs. DeepSeek V4 Pro 0318 vs. Muse Spark 1.2 vs. Grok 4.6 👀
Gave 4 AI models a challenge: Draw an eagle using triangles only 🦅🔺 Gemini 3.7 Flash did really well on this one 👇
You have reached the end of the archive
All of deepseek