Qwen3-VL-235B-A22B-Thinking-FP8 is an open-weight vision-language model supporting text, image, and video inputs. It features advanced reasoning ('thinking') and tool-calling capabilities. The FP8 version offers a balance of performance and efficiency for local or self-hosted inference using the Transformers library.
In the Multimodal & vision space, Qwen3 VL 235B A22B Thinking takes a focused approach. It focuses on providing high-performance multimodal (vision, video, text) reasoning and tool use in an open-weight model. It is built as an open-source project for developers. Qwen3 VL 235B A22B Thinking is open source under the Apache-2.0 license. It runs on the web and the command line.
It is developed by Qwen, and it first shipped in 2024. Development happens publicly on GitHub with 19.6k stars. It operates in a well-populated space: PulseGate tracks 10 similar projects. Key capabilities include vision-language processing, video understanding, and tool calling.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Closest matches by what these projects do