Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Visual Spatial Reasoning on SpaceVista-Bench Outdoor
Loading...
43
Score
GPT-5
10.344
18.822
27.3
35.778
Oct 10, 2025
Score
Updated 7d ago
Evaluation Results
Method
Method
Links
Score
GPT-5
Model Category=Closed-...
2025.10
43
SpaceVista-7b
Model Category=Open-So...
2025.10
39.1
GPT-4o
Model Category=Closed-...
2025.10
38.3
Internvl3-38B
Model Category=Open-So...
2025.10
38
Claude-Sonnet-4
Model Category=Closed-...
2025.10
34.1
Qwen2.5VL-32B
Model Category=Open-So...
2025.10
30.7
Internvl3-78B
Model Category=Open-So...
2025.10
30.3
Claude-Opus-4.1
Model Category=Closed-...
2025.10
30
Gemini-2.5-pro
Model Category=Closed-...
2025.10
29
Qwen2.5VL-72B
Model Category=Open-So...
2025.10
28
Internvl3.5-38B
Model Category=Open-So...
2025.10
27
Gemini-2.5-flash
Model Category=Closed-...
2025.10
26.9
VLM-3R
Model Category=Open-So...
2025.10
26.9
GLM-4.5V
Model Category=Open-So...
2025.10
25.2
Internvl3.5-14B
Model Category=Open-So...
2025.10
24.3
Spatial-MLLM
Model Category=Open-So...
2025.10
23.1
SpaceR
Model Category=Open-So...
2025.10
19.8
GLM-4.1V-Thinking
Model Category=Open-So...
2025.10
13.3
LLAVA-Onevision-72B
Model Category=Open-So...
2025.10
11.7
LLAVA-Onevision-7B
Model Category=Open-So...
2025.10
11.6
Feedback
Search any
task
Search any
task