# Long sequence processing

The latest pre-trained model designed for high-quality code understanding and semantic retrieval, supporting long sequence processing of 8 programming languages.

This is a 4-bit GPTQ quantized version based on the mistralai/Devstral-Small-2505 model, optimized for consumer-grade hardware.

A sentence transformer model specifically designed for code search, based on ModernBERT architecture, supporting sequence lengths of up to 2048 tokens

Isoformer is a multimodal biological sequence model capable of processing DNA, RNA, and protein sequences, and predicting gene expression.

phi-2-super-GGUF is the GGUF quantized version of the abacaj/phi-2-super model, suitable for local execution and text generation tasks.

Mamba is an efficient language model based on the State Space Model (SSM), with the ability to model sequences with linear time complexity.

SauerkrautLM Mixtral 8X7B is a multilingual text generation model based on the Mixtral architecture. It has been fine-tuned and aligned using SFT and DPO, and supports English, German, French, Italian, and Spanish.

Mythalion Kimiko v2 - AWQ is the AWQ quantized version of the Mythalion Kimiko v2 model created by nRuaif, featuring high efficiency, accuracy, and fast inference.

The GPTQ quantized version of Mistral 7B Instruct v0.1, supporting operation under the ExLlama or Transformers framework

CodeFuse-13B is a 13-billion-parameter code generation model trained on the GPT-NeoX framework, supporting over 40 programming languages and capable of processing code sequences up to 4096 characters long.

A Korean text generation model based on the GPT-J architecture with 350 million parameters, suitable for various Korean text generation tasks.

LongT5 is a Transformer-based text-to-text model specifically designed to handle long sequence inputs, supporting up to 16,384 tokens.

Long T5 is a text-to-text Transformer model extended from T5, which supports efficient processing of long sequence inputs and is particularly suitable for text generation tasks.

Funnel Transformer is an English text pre-training model based on self-supervised learning, adopting objectives similar to ELECTRA, achieving efficient language processing by filtering sequence redundancy.

BERTIN is a series of Spanish language models based on BERT. This model follows the RoBERTa-large architecture, trained from scratch using the Flax framework, with data sourced from the Spanish portion of the mC4 corpus.

A character-level Reformer language model trained on the enwik8 dataset, designed for text generation and compression tasks.