MAI-DS-R1 is the result of Microsoft AI team's post-training of the DeepSeek-R1 inference model, aimed at enhancing its response capability to sensitive topics and optimizing risk performance, while maintaining the original reasoning ability and competitive advantages.

Natural Language ProcessingTransformers

MAI-DS-R1 is the result of Microsoft AI team's post-training of the DeepSeek-R1 inference model, aimed at enhancing its response capability to sensitive topics and optimizing risk performance, while maintaining the original reasoning ability and competitive advantages.

Natural Language ProcessingTransformers

Qwen2.5-14B-it-restore is a model generated by combining the della, ties, and model stock fusion methods based on the Qwen2.5-14B and Qwen2.5-14B-Instruct models, aiming to address the decline in instruction-following capability and weakened mathematical ability.

Natural Language ProcessingSafetensorsMultiple Languages

PAVE is a model focused on repairing and adapting video large language models, aiming to enhance the conversion capability between video and text.

MultimodalTransformers

This is a Mixture of Experts model obtained by merging multiple pre-trained language models using mergekit, aiming to create the most unconstrained text generation capability.

Natural Language ProcessingTransformers

Aiden T5 is a groundbreaking transformers model with internet access capability and a Belief-Desire-Intention (BDI) system, combining the powerful capabilities of transformer language models with the ability to learn and reason through the internet and its own cognition.

Natural Language ProcessingTransformersMultiple Languages

This model focuses on the Natural Language Inference (NLI) task, aiming to surpass simple heuristic methods and improve the model's generalization capability in complex reasoning scenarios.

Natural Language ProcessingTransformers