Ep 215: Kolibri activates 3.46 billion parameters per token while landing on the Pareto frontier…

Models & Agents — Video Edition · Oct 26, 2026 · 12m

Models & Agents Kolibri activates 3.46 billion parameters per token while landing on the Pareto frontier for English and German quality versus serving cost. What You Need to Know: Tsinghua released AKV, an SGLang-based cache engineering server that lets LLM agents drop and reposition KV cache instead of summarizing history, cutting average cost 27.7 percent across four agent benchmarks. Kolibri, a new 78.1B-parameter English-German MoE with 3.46B active parameters, was trained on 24T tokens with over 20 percent German data and released as open weights under Apache 2.0. SPECMEM keeps a live…