# LoRA optimization

Flux-Dev2Pro is a fine-tuned model of Flux-Dev's transformer, optimized for LoRA training effectiveness, suitable for image generation tasks.

Flux-Dev2Pro is a Transformer model fine-tuned based on Flux-Dev, optimizing LoRA training performance and restoring Flux-Pro characteristics through multi-step fine-tuning.

A Turkish language model based on ytu-ce-cosmos/turkish-gpt2-medium, fine-tuned with a 35,000-instruction dataset, suitable for various text generation tasks

A Turkish instruction fine-tuned language model based on ytu-ce-cosmos/turkish-gpt2-large, fine-tuned with a dataset of 35,000 instructions

Airavata is a multilingual large language model fine-tuned on the IndicInstruct dataset based on the 7B-parameter OpenHathi model, supporting both English and Hindi.