LIZZY 7B
by Flower
A state-of-the-art LLM built in the UK. Sovereign, secure, and deployment-ready for enterprise and public-sector use.
Designed for the UK
Built from the ground up for UK use cases, standards, and deployment environments.
Tune to internal AI infrastructure
Built from the ground up for UK use cases, standards, and deployment environments.
Incorporate proprietary data
Delivers leading performance among Europe’s top sovereign models.
Align with your compliance and regulatory needs
Training and inference across diverse GPU infrastructure and supports regulated and proprietary data.
Top European Model Performance
Lizzy matches or outperforms existing sovereign European LLMs across a range of downstream benchmarks, including our new benchmark that quantifies key UK relevant metrics.
| Benchmark | Lizzy 7B | EuroLLM 9B | Apertus 8B |
|---|---|---|---|
| Britishness MCQ | 71.0 | 77.6 | 80.8 |
| Britishness CoT | 80.1 | 72.1 | 31.7 |
| Britishness Domains | 89.9 | 69.0 | 32.6 |
| MATH | 77.9 | 31.3 | 22.4 |
| OMEGA | 29.0 | 4.7 | 5.0 |
| BigBenchHard | 69.0 | 38.9 | 42.4 |
| AGI Eval English | 65.6 | 50.2 | 50.4 |
| MMLU | 67.9 | 57.4 | 63.4 |
| GPQA | 34.6 | 26.8 | 28.1 |
Try Lizzy Today
Deploy anywhere with full control. Compatible with vLLM and leading inference stacks. Launch via Hugging Face with one-click.
Enterprise
24/7 support for deployment across your data and compute systems.
Full-stack Decentralized AI
This foundation enables scalable model development without relying on centralized infrastructure and supports Flower's broader vision for sovereign AI.
