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

Hugging Face
VISIONPricing:Input $0.4 / Cached $0.08 / Output $0.8Concurrent Unit Cost:1Model Size:4BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Jul 8, 2026License:apache-2.0Architecture:Transformer0.0K Open Weights Featherless Exclusive Warm

Qwen3-VL-4B-Instruct by Qwen is a 4-billion-parameter dense vision-language model, part of the Qwen3-VL series. It integrates a vision encoder with a dense transformer decoder, utilizing Interleaved-MRoPE positional embeddings and DeepStack multi-level feature fusion. This model excels at visual understanding tasks including OCR, document/chart analysis, spatial reasoning, video comprehension, and natively supports tool calling, processing both images and text inputs to generate text outputs.

Loading preview...

Overview

Qwen3-VL-4B-Instruct is a 4-billion-parameter dense vision-language model from the Qwen3-VL series, developed by Qwen. It combines a vision encoder with a dense transformer decoder, leveraging Interleaved-MRoPE positional embeddings and DeepStack multi-level feature fusion. This model is designed to handle both images and text, making it suitable for a wide range of multimodal tasks. This specific version is the "Instruct" edition, optimized for instruction-following.

Key Capabilities

  • Vision-language processing: Accepts OpenAI-style multimodal chat messages, allowing image_url content parts alongside text for comprehensive visual understanding.
  • Tool calling: Supports tool (function) calling via the hermes tool-call parser, enabling integration with external functionalities.
  • Multimodal understanding: Capable of tasks such as OCR, document and chart analysis, spatial reasoning, and video comprehension.

Good For

  • Applications requiring visual and textual input: Ideal for scenarios where understanding both images and text is crucial, such as analyzing documents with embedded charts or interpreting visual scenes.
  • Developing intelligent agents: Its tool-calling capability makes it suitable for building agents that can interact with external tools based on multimodal prompts.
  • Deployment on FuriosaAI hardware: This model is provided with a Furiosa Executable Bundle (FXB) for optimized performance on FuriosaAI RNGD with Furiosa-LLM, running with a tensor-parallel size of 8 PEs per card.

Finetunes
1 models