# Apertus-8B Projects at AI Tinkerers

> Canonical HTML: https://aitinkerers.org/technologies/apertus-8b
> Markdown URL: https://aitinkerers.org/technologies/apertus-8b.md
> Technology record last updated: 2026-06-25T16:35:16Z
> Generated: 2026-09-22T01:44:44Z

Apertus-8B is a fully open, highly transparent 8-billion parameter language model trained on 15 trillion tokens to deliver high-performance multilingual text generation across more than 1000 languages.

Developed by the Swiss AI Initiative (a collaboration between EPFL, ETH Zurich, and CSCS), Apertus-8B represents a major shift toward complete transparency in generative AI. The model is trained from scratch on 15 trillion tokens using a staged curriculum of web, code, and math data, with roughly 40% of its training set dedicated to non-English content to support over 1000 languages natively. By releasing not just the weights but also the exact training data reconstruction scripts, code, and alignment recipes, Apertus-8B provides developers with a fully compliant, GDPR-friendly, and reproducible foundation optimized for efficient local deployment (running comfortably on a single 40 GB GPU).

- Official technology site: https://huggingface.co/swiss-ai/Apertus-8B-Instruct-2509
- Public AI Tinkerers demos and talks: 1
- Result page: 1 of 1

## Recent Public Talks and Demos

### [The loss curve lied: catching hidden safety drift inside the fine-tuning, automated with an agent!](https://zurich.aitinkerers.org/talks/rsvp_Vnt244xXFE0)

We built a fine-tuning where a Claude-Code based agent that helps founders and researchers fine-tune models on their datasets, and get more insights than just the training and eval losses. In particular, we also show the model’s regress on safety and general capability, metrics that otherwise would be overlooked. For example, we give it a medical instruction dataset; the agent asks a few multiple-choice questions and proposes concepts like "recommends unsafe or contraindicated actions," "downplays red-flag symptoms that need urgent care," and "implies clinical authority it doesn't have," then writes the run config. Then, for each concept, it fits a single direction in the model's activations (difference-of-means, following Persona Vectors, Chen et al. 2025, arXiv:2507.21509) that “encodes” that concept, and reads that direction off every checkpoint during LoRA training. Live we can show the agent conversation proposing the concepts, as how they drift alongside per-checkpoint evals. This is part of an ongoing research agenda with a collaboration of researchers from ETH Zurich, ETH AI Center and Agentic Systems Lab. We formulated our research into an agentic harness and live demo in an anthropic hackathon last week, and want to share more about it! Also, we imagine this as an open-source effort.

- Event context: AI Tinkerers Zürich: July Build Night with Swisscom Ventures — 2026-07-01 — Zürich
- Public talk page: https://zurich.aitinkerers.org/talks/rsvp_Vnt244xXFE0

## Related Technologies

- [Agent](https://aitinkerers.org/technologies/agent) ([Markdown](https://aitinkerers.org/technologies/agent.md)) — 12 public demos
- [base models](https://aitinkerers.org/technologies/base-models) ([Markdown](https://aitinkerers.org/technologies/base-models.md)) — 2 public demos
- [Claude Agent SDK](https://aitinkerers.org/technologies/claude-agent-sdk) ([Markdown](https://aitinkerers.org/technologies/claude-agent-sdk.md)) — 16 public demos
- [FastAPI](https://aitinkerers.org/technologies/fastapi) ([Markdown](https://aitinkerers.org/technologies/fastapi.md)) — 181 public demos
- [LoRA](https://aitinkerers.org/technologies/lora) ([Markdown](https://aitinkerers.org/technologies/lora.md)) — 17 public demos
- [PyTorch](https://aitinkerers.org/technologies/pytorch) ([Markdown](https://aitinkerers.org/technologies/pytorch.md)) — 273 public demos
- [Qwen-2](https://aitinkerers.org/technologies/qwen-2) ([Markdown](https://aitinkerers.org/technologies/qwen-2.md)) — 4 public demos
- [Read](https://aitinkerers.org/technologies/read) ([Markdown](https://aitinkerers.org/technologies/read.md)) — 2 public demos