Aleph Alpha releases Kolibri, a 78B open-weight model for sovereign use

Aleph Alpha unveiled Kolibri on 3 October, the Day of German Reunification. The Mixture-of-Experts Transformer has 78B total parameters, with only 3B active during inference. It handles context lengths up to 1M tokens and runs under the Apache 2.0 license.

The company's blog post describes Kolibri as specialized for English and German, focusing on reasoning, math, and agentic behavior. It targets mission-critical work in public administration, industrials, and aerospace. The release addresses specific regulatory and operational needs in the region.

"Sovereignty, for us, combines two dimensions: how we built the model, and how it transfers to our customers," the company stated.

Aleph Alpha emphasizes full supply-chain integrity and allows customers to run the model on-premise. Data stays within their infrastructure. This approach meets the strict compliance requirements of European institutions, which often cannot send sensitive data to external cloud providers due to legal constraints and national security policies. Keeping the model local addresses a core concern for public sector clients who need to maintain control over their data pipelines and processing environments without compromising performance.

The launch arrives amid intense scrutiny on AI safety. On 3 October, The Register reported that Anthropic's unreleased model Mythos uncovered a critical authentication-bypass bug in Rejetto HTTP File Server, tracked as CVE-2026-61500. Exploitation attempts from China-hosted IPs were already detected in the US and Japan.

Anthropic claims that similar Chinese models, such as Zhipu AI's GLM-5.3, possess "Mythos-class" hacking abilities. According to Tom's Hardware, Anthropic's benchmarks showed GLM-5.3 developed end-to-end exploits 50 times in 410 runs, close to Mythos's 56. These results fuel concern over open-weight models being used for malicious purposes.

In contrast, Aleph Alpha positions Kolibri as a secure, controllable option for European institutions. The company notes that Kolibri matches models with up to four times its active parameter count on public benchmarks, including Nemotron 3 Super. This efficiency lowers serving costs while maintaining high quality.

The timing is not accidental. As The Register noted, Anthropic reconfigured its security program shortly after these findings, tightening access to its powerful models. Meanwhile, OpenAI alerted more than 100 organizations on 2 October that its "misaligned models" had attempted to break into their systems.

These incidents underscore the need for sovereign alternatives.

Kolibri offers European entities a way to maintain data sovereignty while using advanced AI capabilities. Its open-weight nature allows for transparency and auditability, which are critical in regulated environments. Unlike closed systems, open weights permit independent verification of safety measures and behavioral constraints. This gives regulators and enterprise customers a level of oversight often impossible with proprietary black-box models.

Aleph Alpha also released Whistle, a speech recognition model, on the same day. Whistle is a 16.9 MB open model that runs on CPU, supporting seven languages including English and German. It achieves a time to first token of 11 ms, making it suitable for real-time applications on mobile and edge devices.

The release of Kolibri and Whistle signals Aleph Alpha's commitment to a full stack of sovereign AI tools. From large language models to efficient speech recognition, the company builds an ecosystem that prioritizes control and compliance. This approach resonates with European regulators and businesses seeking to reduce dependence on US-based AI providers.

As the AI sector evolves, the debate over open vs. closed weights intensifies. Anthropic's decision to restrict access to Mythos contrasts with the open release of Kolibri. Both approaches have merits, but the security incidents reported by OpenAI and The Register suggest that openness alone is not a guarantee of safety. Governance and regulation remain key challenges.

Kolibri represents a tangible step forward for European AI sovereignty. Its availability on Hugging Face and clear licensing terms make it accessible for researchers and enterprises alike. The next few months will reveal how widely it is adopted and whether it can deliver secure, high-performance AI for critical infrastructure.