Training-free & open-weight FLUX capabilities as one-line Modular Diffusers community pipelines
AI & ML interests
Machine learning, deep learning, generative AI, LLMs
Recent Activity
Test Time Compute for Quantitative Spatial Reasoning using synthetic reasoning traces from 3D scene graphs
-
remyxai/SpaceQwen3-VL-2B-Thinking
Image-Text-to-Text ā¢ 2B ā¢ Updated ā¢ 15 ā¢ 3 -
remyxai/SpaceOm
Image-Text-to-Text ā¢ 4B ā¢ Updated ā¢ 167 ā¢ 13 -
remyxai/SpaceThinker
Viewer ā¢ Updated ā¢ 12.7k ā¢ 220 ā¢ 15 -
remyxai/SpaceThinker-Qwen2.5VL-3B
Image-Text-to-Text ā¢ 4B ā¢ Updated ā¢ 1.02k ā¢ 27
VLMs fine-tuned for spatial VQA using the OpenSpaces dataset.
A collection of tools to automate a variety of tasks, from video processing to solving optimization problems.
Training-free & open-weight FLUX capabilities as one-line Modular Diffusers community pipelines
Off-the-shelf FLUX up to 4096Ā², no fine-tuning where each method a one-line Modular Diffusers pipeline. Swap the repo id to A/B on the same weights.
Test Time Compute for Quantitative Spatial Reasoning using synthetic reasoning traces from 3D scene graphs
-
remyxai/SpaceQwen3-VL-2B-Thinking
Image-Text-to-Text ā¢ 2B ā¢ Updated ā¢ 15 ā¢ 3 -
remyxai/SpaceOm
Image-Text-to-Text ā¢ 4B ā¢ Updated ā¢ 167 ā¢ 13 -
remyxai/SpaceThinker
Viewer ā¢ Updated ā¢ 12.7k ā¢ 220 ā¢ 15 -
remyxai/SpaceThinker-Qwen2.5VL-3B
Image-Text-to-Text ā¢ 4B ā¢ Updated ā¢ 1.02k ā¢ 27
Features VLMs fine-tuned for enhanced spatial reasoning using a synthetic data pipeline similar to Spatial VLM.
VLMs fine-tuned for spatial VQA using the OpenSpaces dataset.
A Collection of Florence-2 fine-tunes
A collection of tools to automate a variety of tasks, from video processing to solving optimization problems.