Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
General Visual Question Answering on HallusionBench
Loading...
69.9
Pass@1
Gemini-3-Pro High
54.612
58.581
62.55
66.519
May 11, 2025
Jul 19, 2025
Sep 26, 2025
Dec 4, 2025
Feb 11, 2026
Apr 21, 2026
Jun 30, 2026
Pass@1
Updated 23d ago
Evaluation Results
Method
Method
Links
Pass@1
Gemini-3-Pro High
2026.06
69.9
Seed2.0 Pro
2026.06
68
GPT-5.2 High
2026.06
67.7
Seed2.0 Lite
2026.06
66
Claude-Opus-4.5
2026.06
65.3
Seed2.0 Mini
2026.06
65.1
Seed1.8
2026.06
63.9
Gemini 1.5 Pro
thinking=true, decodin...
2025.05
63.7
Seed 1.5-VL
thinking=true, decodin...
2025.05
60.3
Seed 1.5-VL
thinking=false, decodi...
2025.05
60
Claude 3.7 Sonnet
thinking=true, decodin...
2025.05
58.3
GPT-4o
thinking=false, decodi...
2025.05
56.2
OpenAI o1
thinking=true, decodin...
2025.05
55.6
Qwen 2.5-VL 72B
thinking=false, decodi...
2025.05
55.2
Feedback
Search any
task
Search any
task