Remyx AI

Enterprise

company

Verified

https://remyx.ai

remyxai

Activity Feed Request to join this org

AI & ML interests

Machine learning, deep learning, generative AI, LLMs

Recent Activity

salma-remyx updated a model about 9 hours ago

remyxai/SpaceThinker-Qwen2.5VL-3B

salma-remyx updated a collection 2 days ago

SpaceThinker

salma-remyx updated a collection 2 days ago

SpaceThinker

View all activity

remyxai's activity

salma-remyx

updated a model about 9 hours ago

remyxai/SpaceThinker-Qwen2.5VL-3B

Image-Text-to-Text • Updated about 9 hours ago • 1.18k • 9

salma-remyx

updated a collection 2 days ago

SpaceThinker

Collection

Test Time Compute for Quantitative Spatial Reasoning using synthetic reasoning traces from 3D scene graphs • 5 items • Updated 2 days ago • 1

salma-remyx

updated a dataset 16 days ago

remyxai/SpaceThinker

Viewer • Updated 16 days ago • 12.7k • 388 • 8

salma-remyx

updated a Space 16 days ago

README

📊

salma-remyx

posted an update 16 days ago

Post

1757

SpaceThinker-Qwen2.5VL-3B shows a 3B VLM can compete with closed, frontier APIs in quantitative spatial reasoning, a key capability for embodied AI applications like drones and robotics.

Check out how it stacks up against Gemini and OpenAI on Q-Spatial-Bench in the ModelCard. Includes .gguf, colab quickstart, docker images.

SpaceThinker adopts the Qwen2.5VL-3B architecture, fine-tuned on the SpaceThinker dataset of synthetic spatial reasoning traces, created with VQASynth

This model builds upon the SpaceLLaVA series of VLMs finetuned for enhanced spatial reasoning using synthetic data by adding test-time compute for multimodal thinking.

Model: remyxai/SpaceThinker-Qwen2.5VL-3B
Dataset: remyxai/SpaceThinker
Space: remyxai/SpaceThinker-Qwen2.5VL-3B
Code: https://github.com/remyxai/VQASynth
Discussion: open-r1/README#10