Lam-3 is a small language model (SLM) in the Lamina series, fine-tuned based on Clemylia/lamina-suite-pretrain. This model can answer questions and is creative, but sometimes the answers may be rather peculiar. The model was created from scratch and trained on the Clem27sey/Nacid dataset containing 1358 question-answer pairs for 3 hours.

Natural Language ProcessingSafetensorsFrench

LAM-1 is the first complete version of the small language model in the lamina series, developed by Clemylia. This model focuses on creative content generation and is optimized to generate imaginative, poetic, and narrative answers rather than factual information.

Natural Language ProcessingTransformersFrench

Tesity-T5 is an encoder-decoder type language model developed by Clemylia, focusing on contextual Q&A tasks. This model can accurately extract and synthesize answers based on the provided text context, with the ability to provide factual and comprehensive responses.

Natural Language ProcessingTransformersFrench

Qwen2.5-VL-7B-Instruct is a multimodal vision-language model developed by the Tongyi Qianwen team at Alibaba. Based on a parameter scale of 7 billion, it is specifically optimized and trained for visual question-answering tasks. This model can understand and analyze image content and generate accurate natural language answers.

MultimodalSafetensorsEnglish

NVIDIA Nemotron Nano 12B v2 is a large language model trained from scratch by NVIDIA, designed for both inference and non-inference tasks. It adopts a hybrid Mamba2-Transformer architecture, supports multiple languages, has controllable inference capabilities, and can generate inference processes or directly give answers according to user needs.

Natural Language ProcessingTransformersMultiple Languages

This is an astronomy professional model fine-tuned based on the Gemma-3-4B model. It is specifically optimized for Portuguese astronomy Q&A and can provide more accurate and contextually relevant Portuguese astronomy answers.

Natural Language ProcessingTransformersEnglish

A technical multiple-choice question generation model fine-tuned based on T5-base, specifically designed to generate relevant questions according to technical context and correct answers, suitable for the creation of educational content and assessments.

Natural Language ProcessingSafetensorsMultiple Languages

TripleBits-Sinhala-Gemma-2B-QnA is a Sinhala Q&A model based on the Google Gemma 2B architecture, fine-tuned using PEFT technology. This model has undergone continuous pre-training on a diverse Sinhala corpus and is specifically optimized for Sinhala Q&A tasks, capable of providing accurate answers for Sinhala Q&A scenarios.

Natural Language ProcessingPeft

A fine-tuned version optimized for providing precise and concise French medical responses. Based on the google/medgemma-4b-it base model, it can give brief (up to 2 - 3 sentences) and medically accurate answers.

Natural Language ProcessingPytorchFrench

RM-R1 is a training framework for reasoning reward models (ReasRM), which evaluates candidate answers by generating scoring criteria or reasoning trajectories, providing interpretable evaluations.

Natural Language ProcessingTransformersEnglish

RM-R1 is a training framework for reasoning reward models (ReasRM), which evaluates candidate answers by generating scoring criteria or reasoning traces, providing explainable judgments.

Natural Language ProcessingTransformersEnglish

RM-R1 is a training framework for inference reward models. It evaluates two candidate answers by generating structured scoring criteria or inference trajectories and provides interpretable reasons.

Natural Language ProcessingTransformersEnglish

RM-R1 is a training framework for reasoning reward models (ReasRM), which evaluates candidate answers by generating scoring criteria or reasoning traces, significantly improving accuracy and interpretability compared to traditional reward models.

Natural Language ProcessingTransformersEnglish

RM-R1 is a training framework for reasoning reward models (ReasRM), which evaluates candidate answers by generating scoring criteria or reasoning traces, providing explainable assessments.

Natural Language ProcessingTransformersEnglish

ELYZA-Shortcut-1.0-Qwen-32B is a non-reasoning model developed based on Qwen2.5-32B-Instruct, capable of bypassing reasoning steps to directly generate final answers.

Natural Language ProcessingTransformersMultiple Languages

A GPT-2 model fine-tuned specifically for Turkish reverse generation tasks, capable of generating corresponding questions from given answers

Natural Language ProcessingTransformersOther

A document QA model fine-tuned from sberbank-ai/ruBert-large, excelling at extracting answers from documents, supporting Russian and English.

Natural Language ProcessingTransformersMultiple Languages

A document question-answering model fine-tuned based on sberbank-ai/ruBert-base, excelling at extracting answers from document contexts, supporting Russian and English.

Natural Language ProcessingTransformersMultiple Languages

A document Q&A specialized model fine-tuned based on ruBert-base, optimized for extracting answers from given document contexts.

Natural Language ProcessingTransformersMultiple Languages

A Q&A pair generation model fine-tuned on T5-base, capable of generating coherent questions even when answers are partially missing

Natural Language ProcessingTransformersEnglish