cyankiwi/diffusiongemma-26B-A4B-it-AWQ-INT4 is a heavily quantized (AWQ INT4) version of a 26B parameter multimodal model based on Gemma. It combines language understanding with diffusion-based generation capabilities and includes an instruction-tuned chat template. The model is intended for local or optimized inference environments.
In the Foundation models & chat space, Diffusiongemma 26B A4B It takes a focused approach. It focuses on running a large multimodal Gemma-based diffusion model in a highly quantized format for local inference. It is built as an open-source project for AI researchers and developers. The project is open source (Open Source). Diffusiongemma 26B A4B It is available on the web, the command line, and API.
Behind Diffusiongemma 26B A4B It is cyankiwi. It operates in a well-populated space: PulseGate tracks 8 similar projects. Among its 3 catalogued features are Multimodal Generation, AWQ Quantization, and Instruction Tuning.
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