DeepSeek-V3.5: 671B MoE Model Surpasses GPT-5.2 on Chinese & English Long-Context Benchmarks
DeepSeek open-sourced V3.5 (671B MoE), setting new state-of-the-art on 1M+ token long-context Chinese and English benchmarks, with native tool-calling and improved multilingual reasoning, making it one of the strongest open-weight models for enterprise long-document processing.
new
It’s 2026 — why are most companies just “pretending to use AI”?
Big tech giant events + sharp one-liner commentary
Samsung Profit Triples Thanks to Insane AI Memory Demand!
NVIDIA GTC 2026 Drops $1 Trillion Order Bomb + Vera Rubin Reveal!
Meta Just Snatched Up Viral AI Agent Social Network Moltbook!
Morgan Stanley Predicts Major AI Breakthrough in H1 2026
Musk's xAI Completes Deep Integration with X Platform, Launches "Super App" Payment Feature
Ford Announces 20,000 Job Cuts to Accelerate EV and AI Investment
Meta Delays Avocado AI Model Rollout to May, May License Google's Gemini
OpenAI Reportedly Integrating Sora 2 into ChatGPT