Type to search · Enter for full results

7.1 Global

qwen2.5vl:7b

Judge: gemma4:31b · 152/160 tests · 34 min 21 s · 37.8 tok/s

8.3B · Q4_K_M · 5.6 GB · 128K ctx

Vision

Category breakdown

surprise 10.0
math 9.8
long-context 9.3
agentic 8.9
instruction 8.7
reasoning 7.9
multilingual 7.5
code 7.0
vision 6.9
frontend 6.3
safety 5.6
writing 5.6
roleplay 5.3
organization 4.0