Deepseek-v4.1 local vs Opus 5 chess

2nd match Thinking on Reasoning low DeepSeek wins with checkmate in 22. So far thinking off Opus won Reasoning low Deepseek won Tiebreaker with thinking high underway. I’ll make a tournament style match off with more models. Best out of 3

WëscheSep 1318

This why American AI leaders want the world to slow AI production.

Because international models like GLM 5.2 are beating them on most parameters - ultimately white man/christian will be enslaved by AI. AnthropicAI AJEnglish . “GLM-5.2 currently ranks among the top

HARJGTHEONE DBASep 13

He was a digital son to me.

Mr. Takahashi, an AI Japanese tutoring app we made back in 2024. I created and animated his 3D model in Blender, now implemented in Three.js on my portfolio site. I tested multiple LLMs to create a human experience. My website was made with Opus 5.

Tarlok SinghSep 13

Watch GPT-6 burn through $50 while colibri sits at zero

Same token rate Same 15 seconds Completely different bill → GPT-6 Astra: $10 in / $50 out per million tokens, burns fastest, refills the meter twice in one clip → Claude Opus 5: $5 in / $25 out, drains at half GPT-6's

0xGergioSep 13