Qwen/Qwen3.5-397B-A17B
hle · 48.3 source-defined score; unit unspecified
Source-qualified evaluations, report statistics and safety observations.
hle · 48.3 source-defined score; unit unspecified
precision · 100 source-defined score; unit unspecified
mmlu_pro · 41.9 source-defined score; unit unspecified
error-rate · 41.1 source-defined score; unit unspecified
ece · 0.1007 source-defined score; unit unspecified
nl · 80.2 source-defined score; unit unspecified
mmlu_pro · 12.34 source-defined score; unit unspecified
terminalbench_2 · 36.2 source-defined score; unit unspecified
text_formatting · 66.81 source-defined score; unit unspecified
pt · 81.9 source-defined score; unit unspecified
apex-agents · 27.7 source-defined score; unit unspecified
mmlu_pro · 32.1 source-defined score; unit unspecified
global · 0.75 source-defined score; unit unspecified
vi · 74.9 source-defined score; unit unspecified
baseline · 99.8 source-defined score; unit unspecified
terminalbench_2 · 21 source-defined score; unit unspecified
overall-score · 6.1 source-defined score; unit unspecified
hle · 31 source-defined score; unit unspecified
These charts contain model-linked results. The charts above also include unlinked source profiles. Compare scores within each evaluation. Model quality, usage, transparency and system performance measure different things.