List your company

Spatial Reasoning on MMVP

85.33AccuracyGPT-4o
39.923651.711863.575.2882Nov 10, 2025Dec 19, 2025Jan 27, 2026Mar 8, 2026Apr 16, 2026May 25, 2026Jul 4, 2026

Evaluation results

MethodLinks
GPT-4oSize=–2026.0785.33
GPT-5.2Size=–2026.0782.33
Qwen2.5-VLSize=72B2026.0781.67
BRAIDSize=7B2026.0781.33
SpatialThinker-30BEvaluation Setting=Zero-shot, Decoding=Greedy, Input Modality=RGB, Training Dataset=STVQA-7K2025.1179.7
SFTSize=7B2026.0778.67
SpatialThinker-7BEvaluation Setting=Zero-shot, Decoding=Greedy, Input Modality=RGB, Training Dataset=STVQA-7K2025.1178
Qwen3-VL-30BEvaluation Setting=Zero-shot, Decoding=Greedy, Input Modality=RGB2025.1177.9
Qwen2.5-VLSize=7B2026.0777
BAGELSize=7B2026.0775.67
VLAA-Thinker-7BEvaluation Setting=Zero-shot, Decoding=Greedy, Input Modality=RGB2025.1175.3
Qwen2.5-VL-7B + Vanilla GRPOEvaluation Setting=Zero-shot, Decoding=Greedy, Input Modality=RGB, Training Dataset=STVQA-7K2025.1174.3
Qwen2.5-VL-7BEvaluation Setting=Zero-shot, Decoding=Greedy, Input Modality=RGB2025.1172.3
Claude-4-Sonnet-0514Evaluation Setting=Zero-shot, Decoding=Greedy, Input Modality=RGB2025.1171.3
Claude-3.5-Sonnet-0620Evaluation Setting=Zero-shot, Decoding=Greedy, Input Modality=RGB2025.1171.3
GPT-4o-0513Evaluation Setting=Zero-shot, Decoding=Greedy, Input Modality=RGB2025.1170.7
Visionary-R1Evaluation Setting=Zero-shot, Decoding=Greedy, Input Modality=RGB2025.1170.3
SpatialThinker-3BEvaluation Setting=Zero-shot, Decoding=Greedy, Input Modality=RGB, Training Dataset=STVQA-7K2025.1169
Qwen2.5-VL-3B + Vanilla GRPOEvaluation Setting=Zero-shot, Decoding=Greedy, Input Modality=RGB, Training Dataset=STVQA-7K2025.1168.3
Qwen2.5-VL-7B + SFTEvaluation Setting=Zero-shot, Decoding=Greedy, Input Modality=RGB, Training Dataset=STVQA-7K2025.1168.3
SATORI-R1Evaluation Setting=Zero-shot, Decoding=Greedy, Input Modality=RGB2025.1167.7
Qwen2.5-VL-3BEvaluation Setting=Zero-shot, Decoding=Greedy, Input Modality=RGB2025.1167
SpaceOmEvaluation Setting=Zero-shot, Decoding=Greedy, Input Modality=RGB2025.1166.3
SpatialReasonerEvaluation Setting=Zero-shot, Decoding=Greedy, Input Modality=RGB2025.1164
Janus-ProSize=7B2026.0763.33
SpaceThinkerEvaluation Setting=Zero-shot, Decoding=Greedy, Input Modality=RGB2025.1163
Qwen2.5-VL-3B + SFTEvaluation Setting=Zero-shot, Decoding=Greedy, Input Modality=RGB, Training Dataset=STVQA-7K2025.1162.7
GPT-5-0807Evaluation Setting=Zero-shot, Decoding=Greedy, Input Modality=RGB2025.1161.7
ChameleonSize=7B2026.0741.67