Cohere Releases North Small Translate: A 218B MoE Translation Model That Scores 83.6 on WMT26 Across 50 Languages
AI News

Cohere Releases North Small Translate: A 218B MoE Translation Model That Scores 83.6 on WMT26 Across 50 Languages

Cohere has released North Small Translate, an open-weight machine translation model from Cohere and Cohere Labs. It is a sparse Mixture-of-Experts (MoE) model with 218B total and 25B active parameters. It covers 50 languages, from […]

How AI agents fix it
AI News

How AI agents fix it

Supply chain disruption cost businesses about $184 billion in 2025, according to the J.S. Held Global Risk Report, and most of that bill still buys faster detection, not faster action. That figure is usually treated […]

DeepSeek AI Released DeepSeek-V4.1-Flash with 1M Context, FP4 KV Cache, and Cross-Layer Attention Reuse
AI News

DeepSeek AI Released DeepSeek-V4.1-Flash with 1M Context, FP4 KV Cache, and Cross-Layer Attention Reuse

Long-horizon agents have turned LLM serving into an input-heavy workload. Repeated prefills and million-token contexts leave KV caches that strain HBM, SSD capacity, and bandwidth. DeepSeek AI built its newest release around that exact bottleneck. […]