Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Complex Reasoning on TVbench
Loading...
54.7
Accuracy
Qwen3-VL-8B + SynRL
51.788
52.544
53.3
54.056
Mar 18, 2026
Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
Qwen3-VL-8B + SynRL
Model Scale=8B, SynRL...
2026.03
54.7
Qwen3-VL-8B
Model Scale=8B, SynRL...
2026.03
54
Qwen3-VL-4B + SynRL
Model Scale=4B, SynRL...
2026.03
53.6
Qwen3-VL-4B
Model Scale=4B, SynRL...
2026.03
51.9
Feedback
Search any
task
Search any
task