furiosa-ai/Qwen3-VL-4B-Instruct

Hugging Face
VISIONConcurrent Unit Cost:1Model Size:4BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Jul 8, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Warm

Qwen3-VL-4B-Instruct by Qwen is a 4-billion-parameter dense vision-language model from the Qwen3-VL series. It integrates a vision encoder with a dense transformer decoder, utilizing Interleaved-MRoPE and DeepStack for multimodal input processing. This model excels at visual understanding tasks including OCR, document analysis, spatial reasoning, and video comprehension, and natively supports tool calling. It is designed for instruction-following applications, processing both image and text inputs to generate text outputs.

Loading preview...

Overview of Qwen3-VL-4B-Instruct

Qwen3-VL-4B-Instruct is a 4-billion-parameter dense vision-language model developed by Qwen. It is part of the Qwen3-VL series, designed to process and understand both visual and textual information. The model employs a vision encoder paired with a dense transformer decoder, leveraging Interleaved-MRoPE positional embeddings and DeepStack multi-level feature fusion to effectively handle images and videos alongside text.

Key Capabilities

  • Multimodal Input: Accepts OpenAI-style multimodal chat messages, allowing image_url content parts alongside text.
  • Comprehensive Visual Understanding: Capable of performing tasks such as OCR, analysis of documents and charts, spatial reasoning, and video comprehension.
  • Native Tool Calling: Supports tool (function) calling through the hermes tool-call parser, enabling interaction with external functions.
  • Instruction Following: This is the "Instruct" edition, optimized for following instructions in multimodal contexts.

Good For

  • Applications requiring visual and textual input: Ideal for scenarios where understanding both images and text is crucial.
  • Multimodal chat interfaces: Can be integrated into chat systems that need to interpret and respond to visual queries.
  • Automated document and image analysis: Suitable for tasks involving extracting information or insights from visual content.
  • Tool-augmented AI systems: Its native tool-calling support makes it effective for building agents that can interact with external tools based on multimodal prompts.

Finetunes
1 models