Collections

DeepSeek R1 - the most powerful reasoning open-source model - available in GGUF, original & 4-bit formats. Includes Llama & Qwen distilled models.

GemmaX2 language models, including pretrained and instruction-tuned models of 2 sizes, including 2B, 9B.

Trained based on Flux-generated images and their enhanced prompt captioning and annotations in a simple JSON format.