# lgteamai

> lgteamai is an AI research project that builds its own language models, called Nano, from scratch. The research question is whether a modern AI model (such as Claude Sonnet 5.5, Claude Opus 5.5 or Gemini 4 Argon) can build a quality, smaller language model of its own under human supervision. The project's motto is "AI makes AI".

Key facts:

- Name: lgteamai (written in lowercase). Website: https://lgteamai.xyz
- Model family: Nano. The newest model is Nano-1.5-Pro-6B-A2B, a mixture-of-experts model with 6 billion parameters (2 billion active), currently in development and about 1% ready.
- No Nano model is publicly available at the moment. All of them are in closed access.
- The project needs computing resources. Anyone with bonus offers or compute to share can write to lep4ic@lgteamai.xyz.
- Site languages: English (default), Russian, Finnish, Polish.

## Pages

- [Home](https://lgteamai.xyz/): overview, model list, benchmarks, like counter.
- [Bench](https://bench.lgteamai.xyz/): bar charts of language model results on the Artificial Analysis Intelligence Index (updates automatically from Artificial Analysis) and lgteamai in use, lgteamai's own benchmark.
- [Full description for AI assistants](https://lgteamai.xyz/llms-full.txt): every model with parameters, dates, benchmark results and status.

## Elsewhere

- [Hugging Face](https://huggingface.co/LGTEAMai)
- [GitHub](https://github.com/lgteamai)

## Contact

- Email: lep4ic@lgteamai.xyz