LLM360 is excited to announce several new releases to further our mission enabling community-owned AGI by creating standards and tools to advance the bleeding edge of LLM capability and empower knowledge transfer, research, and development.
K2-65B
The first fully reproducible large language model to outperform Llama 2 70B, using 35% less compute. Trained in two stages, K2-65B has demonstrated on par reasoning and text generation capabilities with strong domain knowledge in medicine, coding, and math. Learn more here and see our technical report here.
The LLM360 Model Performance and Evaluation Collection
A robust large language model evaluation suite consisting of general and domain specific evaluations to assess model knowledge and function. Learn more here.
Please reach out with any feedback or questions to team@llm360.ai
Thanks,
The LLM360 Team