2 comments
senkoabout 1 hour ago
Since it's a PDF, here's an excerpt from the summary:
> This report introduces Kolibri, an English–German Mixture-of-Experts transformer with 78.1B total parameters and 3.46B active parameters per token, released as open weights under the Apache 2.0 licence.
> We built Kolibri for sovereign and specialised deployment, with a particular focus on German and on regulated domains such as public administration, industry, and aerospace. To support these settings, we control the full model-development process, including data, architecture, training infrastructure, post-training, and evaluation.
yu3zhou4about 1 hour ago
It’s not a dupe, it’s a full training and infra report
Read the full thread on Hacker News →
Related stories
- Can you forget how you feel about Meta?theverge.comThe Verge · 0 points · 11 days ago
- Practical Decryption exFiltration: Breaking PDF Encryptionpdf-insecurity.orgLobsters · 3 points · about 7 years ago
- Show HN: Why is my PDF so big? – See every byte of a PDF as a treemapwhyismypdfsobig.comHacker News · 2 points · 9 days ago
- How Big Tech Runs Tech Projects and the Curious Absence of Scrum (2021)newsletter.pragmaticengineer.comHacker News · 1 points · 13 days ago
- How Big Tech Runs Tech Projects and the Curious Absence of Scrum (2021)newsletter.pragmaticengineer.comHacker News · 1 points · 13 days ago
- Lobsters · 9 points · about 4 years ago