Since 2020, LightOn is focused on building state-of-the-art Large Language Models for customers as well as for internal use. Recently LightOn released Alfred-40B-0723 and Alfred-40b-1023, model finetuned with RLHF and with longer context length, based on Falcon-40B. LightOn offers a platform called Paradigm, designed to help large companies navigate the complexities of Generative AI.
- LightOn's open-source Language Models (LLM) based on Falcon-40B.
- Purpose: Designed to integrate Generative AI into your workflow.
- Capabilities: Excels in prompt engineering, no-code application development, and classic LLM tasks.
- Training: Finetuned with Reinforcement Learning from Human Feedback.
- Availability: Accessible on HuggingFace.
Technical details and model weights are available at our HuggingFace model cards: alfred-40b-0723 and alfred-40b-1023.
LightOn is hiring! We are always looking for talented individuals to join our team and help us build the Generative AI Copilot.
We are not credentials-driven. We want to understand your unique background and what you can bring to the team. You should be eligible to work in the EU, but all exceptional profiles will be considered.
