nightbloom/Qwen3-VL-8B-Instruct_AL-RUCaption

VISIONConcurrent Unit Cost:1Model Size:8BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Jan 11, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

The nightbloom/Qwen3-VL-8B-Instruct_AL-RUCaption is an 8 billion parameter vision-language instruction-tuned model, developed by nightbloom and fine-tuned from prithivMLmods/Qwen3-VL-8B-Abliterated-Caption-it. This model is optimized for visual understanding and captioning tasks, leveraging a 32768 token context length. It was trained using Unsloth and Huggingface's TRL library for accelerated performance.

Loading preview...

Model Overview

The nightbloom/Qwen3-VL-8B-Instruct_AL-RUCaption is an 8 billion parameter vision-language model, developed by nightbloom. It is an instruction-tuned variant, building upon the prithivMLmods/Qwen3-VL-8B-Abliterated-Caption-it base model.

Key Characteristics

  • Vision-Language Capabilities: Designed to process and understand both visual and textual inputs, making it suitable for tasks requiring multimodal reasoning.
  • Instruction-Tuned: Optimized to follow instructions effectively, enhancing its utility for various downstream applications.
  • Efficient Training: The model was fine-tuned using Unsloth and Huggingface's TRL library, enabling faster training times.
  • Context Length: Supports a substantial context length of 32768 tokens, allowing for processing of longer inputs.

Use Cases

This model is particularly well-suited for applications involving:

  • Image captioning and description generation.
  • Visual question answering (VQA).
  • Multimodal instruction following.
  • Tasks requiring a blend of visual understanding and natural language generation.