A heavily quantized (4-bit AWQ) version of a Qwen 3.6 model with 35B+3B parameters and vision capabilities (VRAP). The 21.2GB model supports both text and image inputs. It is designed for local inference using GGUF-compatible tools or optimized runtimes.
In the Multimodal & vision space, Qwen 3.6 35B A3B VRAP 4 Bit AWQ 21.2GB takes a focused approach. Efficiently running a large multimodal Qwen model with vision and text understanding on limited hardware. Qwen 3.6 35B A3B VRAP 4 Bit AWQ 21.2GB is an open-source project aimed at developers. The project is open source (Open Source). It runs on the command line.
It is developed by selode-ai, and it first shipped in 2025. The category is crowded — PulseGate's index counts 25 comparable projects. Among its 3 catalogued features are AWQ Quantization, Vision Support, and 4-bit Precision. It exposes integrations via a public API.
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