Aleph Alpha open-sources Kolibri: 78.1B MoE with 3.46B active params under Apache 2.0
petrusenko_max · x · 2026-10-04
Aleph Alpha's Kolibri is now out under Apache 2.0: a German/English mixture-of-experts model with 78.1B total parameters and 3.46B active per token (4.4%).
It was trained from scratch on 24T tokens — over a fifth German — using 768 NVIDIA B200s hosted in Germany and Finland, making it a flagship open-source release in Europe's sovereign AI push.
More from Models
- Daniel Han publishes summary of LLM benchmarks you can actually trust — danielhanchen · 2026-10-06
- Claim Verification Benchmarks Mostly Test Retrieval, Not Reasoning, Finds 24K-Trace Study — deliprao · 2026-10-06
- COLM26 study: LLMs ace claim verification benchmarks by taking shortcuts, not verifying — deliprao · 2026-10-06
- Opus 5.5 uses 26k tokens vs Astra's 12k yet costs 23% less per task at equal AA score — ChrisGPT · 2026-10-06
- GPT-6 Astra claimed to be first AI crossing world-class astrophysics threshold — johnseach · 2026-10-06
- $500/mo AI subscription is huge money in Jakarta: PPP pricing debate — sujingshen · 2026-10-06