FRAMEWIREIndonesiaUpdated Aug 15Live wire

0:00 / 0:00

I ran the same 41 tasks twice on DeepSeek V4-Pro.

The only thing I changed was one word. DeepSeek shipped V4-Pro-0813 on Thursday with a reasoning-effort switch: low, high, max. Same model, same endpoint, same prompt. Everyone reposted the spec sheet. Nobody priced the switch.

BountyAug 151

Keep scrolling for the next clip ↓All of deepseek Archive

0:00 / 0:00

New deepseek v4-pro on 3D scrabble

Pierre TCHIOAug 15

Keep scrolling for the next clip ↓All of deepseek Archive

0:00 / 0:00

Today we celebrate the launch and open sourcing of Qwen 3.8 27b which is a nice step forward for local AI.

For its size (~15GB on disk, 4bit quant) it is unmatched for agentic work on your personal computer. And it's ability to understand images and videos is refreshing. But...

Mau “Rules without Rulers” LedfordAug 156

Keep scrolling for the next clip ↓All of deepseek Archive

0:00 / 0:00

Project Number 11- DeepSeek Harness🔥

'Plugin-based open-source agent harness from DeepSeek'

ManuAGI 🤖 - ( ManuIn )Aug 15

Keep scrolling for the next clip ↓All of deepseek Archive