Holy sht, it can't be fcking real.
DeepSeek V4.1 Flash just hit 98% of GPT-6 Astra's score on real design tasks, at a fraction of the cost. and it's free on Xplabs. worth being precise on that number: it's 98% on design work specifically, not "basically the same model" across
Opus 5 vs DeepSeek v4.1 Flash
Tested both models with same frontend prompt at highest reasoning available but results came out really different opus took 80 minutes to complete and costed $20 v4.1 flash took 60 minutes and costed $0.2 which one did better here?
#Trump👀 🌏⬆️ #Hamas #Hezbollah #Syria #Lebanon #Iraq #Iran ☢️ 🇺🇸👁️🇨🇳 #Israel #UkraineRussianWar…
#Trump👀 🌏⬆️ #Hamas #Hezbollah #Syria #Lebanon #Iraq #Iran ☢️ 🇺🇸👁️🇨🇳 #Israel #UkraineRussianWar #Elon Tech &🤖#WW3 🇺🇸 🇨🇳 #IphoneDuo #Xiaomi18 #Huawei as ☪️🙈💥endlessly!
Really cooked with this one!
Here is a Three.js open world that DeepSeek-V4.1-Flash made for me. One-Shot! Full procedural, no external assets! Single html file! Yah it's a massive improvement over last generation and even when compared to other closed models.
You have reached the end of the archive
All of deepseek