Telah memperluas batas sesi pengkodean deepseek-v4.1-flash menjadi 5 juta token, yang dicapai dengan kompresi konteks waktu inferensi.
Dengan upaya berpikir maksimal, mencapai akurasi 70%+ di DeepSWE. - pada Upaya berpikir rendah, secara signifikan mengungguli GLM-5.2 dan menggunakan 30% lebih sedikit
#Deepseek API is back online!
Now running #Deepseek in OpenAI #Codex harness
This week, China's distillation reckoning was a welcome vindication for PatrickMoorhead.
Patrick says he got dragged online for calling Kimi K3 a distillation of frontier models. But the NSA, CISA, and FBI just proved him right, naming DeepSeek, Moonshot AI, Alibaba, MiniMax,
Compressed textures from 20gb ~> 8gb and this still looks miles better than the iOS version
IOS version is like 25gb and mine is only 13gb I used DeepSeek V4.1 Flash and Hermes Agent to do the job and now I am Running Resident Evil 7 on a midrange android phone smoothly 😁
Sudah sampai ujung arsip
Semua deepseek