A quantized version of the Qwen2.5-VL-3B-Instruct model optimized with AWQ. It accepts both image and video inputs along with text and follows natural language instructions for vision-language tasks. The model is hosted on Hugging Face and can be loaded via the transformers library or run with inference engines supporting GGUF/AWQ formats.
Qwen2.5 VL 3B Instruct sits in PulseGate's Foundation models & chat category. It focuses on running efficient multimodal (vision + language) models locally with reduced memory requirements. It is built as an open-source project for developers and researchers. Qwen2.5 VL 3B Instruct is open source under the Apache-2.0 license. The product ships for the web, the command line, and API.
Qwen builds and maintains Qwen2.5 VL 3B Instruct, and the product first shipped in 2024. Development happens publicly on GitHub with 19.6k stars. The category is crowded — PulseGate's index counts 25 comparable apps. Key capabilities include vision-Language, Instruction Following, and AWQ Quantized.
Latest indexed changes and source events
Qwen/Qwen2.5-VL-3B-Instruct-AWQ verified by the PulseGate indexer
Other apps tracked under the same category.