jordanpainter/diallm-gemma-grpo-all
Image-Text-to-Text • 4B • Updated • 18
DialLM GRPO
Group Relative Policy Optimization fine-tunes for DialLM across Gemma, Llama, and Qwen models, covering all dialect variants.
-
-
jordanpainter/diallm-gemma-grpo-aus
Image-Text-to-Text • 4B • Updated • 10 -
jordanpainter/diallm-gemma-grpo-brit
Image-Text-to-Text • 4B • Updated • 18 -
surrey-nlp/diallm-gemma-grpo-ind
Image-Text-to-Text • 4B • Updated • 17 -
jordanpainter/diallm-llama-grpo-all
Text Generation • 8B • Updated • 9 • 1 -
jordanpainter/diallm-llama-grpo-aus
Text Generation • 8B • Updated • 9 -
jordanpainter/diallm-llama-grpo-brit
Text Generation • 8B • Updated • 7 -
jordanpainter/diallm-llama-grpo-ind
Text Generation • 8B • Updated • 7 -
jordanpainter/diallm-qwen-grpo-all
Text Generation • 8B • Updated • 8 • 1 -
surrey-nlp/diallm-qwen-grpo-aus
Text Generation • 8B • Updated • 173 -
jordanpainter/diallm-qwen-grpo-brit
Text Generation • 8B • Updated • 6 -
jordanpainter/diallm-qwen-grpo-ind
Text Generation • 8B • Updated • 7