Aleph Alpha Releases Kolibri: A 78.1B Open-Weight English-German MoE Model With Only 3.46B Active Parameters
Aleph Alpha released Kolibri, a 78.1B open-weight English-German MoE with 3.46B active parameters.
Aleph Alpha released Kolibri, a 78.1-billion-parameter Mixture-of-Experts model for English and German that activates 3.46 billion parameters per token. It offers a 1-million-token context window and per-request reasoning effort. Apache 2.0 FP8 weights are published and can run on a single NVIDIA B200 or H200.
- 78.1B MoE activates only 3.46B parameters per token.
- Context window is 1 million tokens with adjustable reasoning effort.
- Apache 2.0 FP8 weights fit a single B200 or H200.
Aleph Alpha has released Kolibri, a 78.1B-parameter English-German Mixture-of-Experts model that activates only 3.46B parameters per token. It has a 1M-token context and per-request reasoning effort, and its Apache 2.0 FP8 weights run on a single B200 or H200. The post Aleph Alpha Releases Kolibri: A 78.1B Open-Weight English-German MoE Model With Only 3.46B Active Parameters appeared first on MarkTechPost.
The full text could not be extracted from this site (paywall, bot protection or heavy scripting). Read it at marktechpost.com.