Open models - Language models
Whether you’re looking to study the science of language models or improve how multimodal models interpret our world, we have you covered. Then see how these models stack up using our range of evaluation frameworks.
Featured model - OLMo 2
Meet OLMo 2, the best fully open language model to date, including a family of 7B and 13B models trained up to 5T tokens. OLMo 2 outperforms other fully open models and competes with open-weight models like Llama 3.1 8B.
Tülu 3
Tülu 3 is a leading instruction following model family, offering fully open-source data, code, and recipes designed to serve as a comprehensive guide for modern post-training techniques.
OLMoE
OLMoE is the first mixture-of-experts model to join the OLMo family. OLMoE is the first model to be on the Pareto frontier of performance and size, while also being released with open data, code, evaluations, logs, and intermediate training checkpoints. OLMoE can be trained 2x faster than equivalent dense models.