Aleph Alpha model releases

Aleph Alpha is a German AI company (founded 2019, headquartered in Heidelberg) building sovereign, EU-compliant AI, earlier known for the Luminous and Pharia model families. It now ships open weights as its flagship strategy. First tracked model is Kolibri-1 (October 2026), an Apache-2.0 open-weight bilingual German/English Mixture-of-Experts text model (78.1B total / 3.46B active) with a 1M-token context, built and trained in Germany and Finland. This page collects the lab's model releases, lifecycle events, source links, and model metadata in one crawlable record.

Models

Labs

Open

Recent

All models·Latest releases·Open-weight models·Changelog·Home

Most recent in this set

Oct 3, 2026

Kolibri-1

Aleph Alpha

Sort

1 model

Kolibri-1

Available

Aleph AlphaOpen source

Kolibri-1 is Aleph Alpha's sovereign open-weight text LLM, released on 2026-10-03 under the Apache-2.0 license with weights on Hugging Face (an FP8 flagship "Kolibri-1" plus a BF16 base "Kolibri-1-BF16" and several quantized variants). It is a Mixture-of-Experts model with 78.1B total (78,103,074,560) / ~3.46B active (3,457,573,120) parameters — 50 layers, 384 routed experts, 6 active per token — tuned for bilingual German/English use and trained natively to a 262,144-token context, extensible to 1,048,576 tokens (1M). Text in / text out, with training for grounding, tool-calling and abstention when evidence is insufficient; built and trained in Germany and Finland. Aleph Alpha reports post-training scores of 75.5% overall (English) and 70.8% (German), 96.5% math and 89.3% code (English averages), and 63.2% on RULER at 1M-token context. Knowledge cutoff 2026-06-18; figures are vendor / self-reported.

MoE78.1B1.0M ctxOct 3, 2026