# INT8 Quantization

BAGEL is an open-source 7B active parameter multimodal foundation model supporting multimodal understanding and generation tasks

This is the INT8 quantized version of the Meta-Llama-3.1-8B-Instruct model, optimized through weight and activation quantization, suitable for multilingual business and research applications.

INT8 quantized version of QWQ-32B, optimized by reducing the bit-width of weights and activations

Quantized version of Qwen2.5-VL-7B-Instruct, supporting vision-text input and text output, optimized for inference efficiency through INT8 weight quantization

The quantized version of DeepSeek-R1-Distill-Qwen-14B, optimized with INT8 quantization for weights and activations, reducing GPU memory requirements and improving computational efficiency.

FLUX.1-dev is a text-to-image diffusion model quantized to INT8 format using Optimum Quanto, suitable for non-commercial use.

Quantized (INT8) ONNX variant of BGE-large-en-v1.5 with inference acceleration via DeepSparse

This is the ONNX version of the RoBERTa-base-go_emotions model, supporting full precision and INT8 quantization for multi-label emotion analysis tasks.

This is the INT8 quantized version of the DistilBERT base uncased model, specifically designed for question answering tasks, optimized for model size and inference speed through post-training static quantization.

BERT-large English Q&A model pre-trained with whole word masking and fine-tuned on SQuAD v1.1, quantized to INT8 precision