Stockmark-2-VL-100B-beta is a Japanese-specific vision-language model with 100 billion parameters, equipped with chain-of-thought (CoT) reasoning ability and can be used for document reading and comprehension.

MultimodalTransformersMultiple Languages

A high-performance, multi-domain AI model built using MergeKit's Model Stock fusion technology, integrating multiple fine-tuned Qwen3-4B models, demonstrating exceptional reasoning, programming, and multi-step problem-solving capabilities in structured output and technical application scenarios.

Natural Language ProcessingTransformersEnglish

A text summarization model fine-tuned based on the T5-small architecture, specifically designed for processing stock news content

Natural Language ProcessingTransformers

A 14B-parameter code generation and comprehension model based on the Qwen2.5 architecture, integrated through the Model Stock method by combining multiple specialized coding models

Natural Language ProcessingTransformers

This is a multi-model fusion result based on the Qwen2.5-14B architecture, utilizing the Model Stock fusion method, combining 22 different 14B-parameter scale models from various sources.

Natural Language ProcessingTransformers

A deep learning-based investment prediction system utilizing Transformer architecture, integrating DeepSeep-V3 and LLama3 design structures for stock price trend forecasting and technical analysis.

FinSeer StockLLM is an open-source 1-billion-parameter large language model specifically designed for financial time series forecasting, utilizing a Retrieval-Augmented Generation (RAG) framework.

Natural Language ProcessingTransformersEnglish

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

A merged model based on multiple Qwen series 32B parameter models, enhanced with Model Stock method for improved multilingual processing capabilities

Natural Language ProcessingTransformers

Stockmark-2-100B is a 100-billion parameter large language model focused on Japanese capabilities, pre-trained on 1.5 trillion tokens of multilingual data and enhanced with Japanese synthetic data for improved instruction following.

Natural Language ProcessingTransformersMultiple Languages

A Llama 3.1 8B series model fused using the Model Stock method, combining the strengths of the Dolermed and Smarteaz variants

Natural Language ProcessingTransformers

A news analysis model fine-tuned based on Llama-3.2-3B, supporting summarization, sentiment analysis, stock ticker extraction, and ad recognition

Natural Language ProcessingTransformersKorean

inferor 0.1 is a merged language model based on the Mistral-Nemo architecture. It is a new version created by integrating multiple pre-trained models using the Model Stock merging method. This model performs excellently in text generation and understanding tasks and supports various natural language processing applications.

Natural Language ProcessingTransformers

Inferor is a large language model with 12B parameters obtained by merging multiple pre-trained language models. It uses the advanced Model Stock merging method, combines the advantages of four high-quality basic models, and provides more powerful language processing capabilities and performance.

Natural Language ProcessingTransformers

IndicFinGPT is a professional Transformer model specifically tailored for the Indian financial market. It is redesigned based on the GPT - Neo - 125M architecture and can provide in - depth insights for the NIFTY50 index and the top 100 companies on the BSE and NSE exchanges, meeting the unique needs of investors in the Indian stock market.

Natural Language ProcessingPeftEnglish

A multilingual large language model based on Qwen2.5-7B, integrating code, math, and general instruction-following capabilities

Natural Language ProcessingTransformersMultiple Languages

beLIEve is a text-to-image model that combines 'I Can't Believe It's Not Photography' v1 and 'Real Stock Photos' v3, designed to enhance detail representation and facial diversity.

Computer VisionDiffusersEnglish

Stockmark-100B is a large language model with 100 billion parameters developed by Stockmark. It is pre-trained from scratch based on approximately 910 billion Japanese and English corpus labels. This model focuses on business applications and performs excellently in handling Japanese business problems.

Natural Language ProcessingTransformersMultiple Languages

A merged model based on Llama-3-8b, integrating multiple pre-trained language models using the Model Stock method

Natural Language ProcessingTransformers

SSAF-FinBert is a sentiment analysis model for stock market news, fine-tuned from FinBert, capable of classifying financial texts into positive, negative, and neutral sentiments.

Natural Language ProcessingTransformers