remyxai/SpaceOm
Image-Text-to-Text
•
4B
•
Updated
•
716
•
6
AI & ML interests
Machine learning, deep learning, generative AI, LLMs
Recent Activity
View all activity
salma-remyx
updated
a
model
5 minutes ago
salma-remyx
updated
a
model
42 minutes ago
salma-remyx
updated
a
model
7 days ago
Performance
3
#1 opened 3 months ago
by
yveeckh
## 🚨License Conflict: Apache-2.0 vs Qwen Research License
1
#4 opened 8 days ago
by
xixi126
salma-remyx
updated
a
Space
13 days ago
salma-remyx
posted
an
update
14 days ago
Post
1553
When multiple benchmarks yield conflicting model rankings, how do you know which model to trust?
In this substack, we explore that question in the context of spatial reasoning capabilities as seen from the perspective of 3 new benchmarks.
Read more: https://remyxai.substack.com/p/benchmark-fusion
In this substack, we explore that question in the context of spatial reasoning capabilities as seen from the perspective of 3 new benchmarks.
Read more: https://remyxai.substack.com/p/benchmark-fusion
salma-remyx
posted
an
update
2 months ago
Post
1862
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
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
-
- 1 reply
salma-remyx
updated
a
model
2 months ago
salma-remyx
updated
a
dataset
2 months ago
salma-remyx
updated
3
models
2 months ago