List your company

Visual Spatial Intelligence Reasoning on VSI-Debiased

0.699AccuracySSR-3D
0.203960.332480.4610.58952Feb 5, 2026Feb 8, 2026Feb 12, 2026Feb 16, 2026Feb 20, 2026Feb 24, 2026Feb 28, 2026

Evaluation results

MethodLinks
SSR-3DCategory=Open-Sourced spatial models2026.020.699
GeoThinker Qwen3vl-8B-32frame# Frames=1282026.020.681
GeoThinker Qwen3vl-8B-32frame# Frames=642026.020.677
SSR-2DCategory=Open-Sourced spatial models2026.020.666
GeoThinker Qwen3vl-8B-32frame# Frames=322026.020.663
GeoThinker Qwen3vl-8B-8frame# Frames=1282026.020.653
GeoThinker Qwen3vl-8B-8frame# Frames=322026.020.648
GeoThinker Qwen3vl-8B-8frame# Frames=642026.020.643
GeoThinker Qwen3vl-8B-32frame# Frames=162026.020.643
SenseNova-SI InternVL3-8B# Frames=322026.020.628
SenseNova-SI (InternVL3-8B)Category=Open-Sourced spatial models2026.020.628
SenseNova-SI InternVL3-8B# Frames=642026.020.624
GeoThinker Qwen3vl-8B-8frame# Frames=162026.020.607
Cambrian-S-7B# Frames=1282026.020.599
SenseNova-SI InternVL3-8B# Frames=1282026.020.597
Cambrian-S-7B# Frames=642026.020.591
SenseNova-SI InternVL3-8B# Frames=162026.020.589
Cambrian-S-7BCategory=Open-Sourced spatial models2026.020.563
Cambrian-S-7B# Frames=322026.020.556
VG-LLM-8B*# Frames=642026.020.552
VG-LLM-8B*# Frames=1282026.020.551
VG-LLM-8B*# Frames=322026.020.524
VG-LLM-8B*# Frames=162026.020.516
Cambrian-S-7B# Frames=162026.020.497
LLaVA-Video-7BCategory=Open-Sourced general models2026.020.307
LLaVA-OneVision-7BCategory=Open-Sourced general models2026.020.285
SmolVLM2-2.2BCategory=Open-Sourced general models2026.020.223