deepseek-ai/DeepSeek-V4.1-Flash-1017
terminalbench_4 · 42.3 source-defined score; unit unspecified
Source-qualified evaluations, report statistics and safety observations.
terminalbench_4 · 42.3 source-defined score; unit unspecified
swe_atlas_qna · 49.5 source-defined score; unit unspecified
text_content · 83.7 source-defined score; unit unspecified
ale-cli-pass-rate · 23.809523809523807 source-defined score; unit unspecified
diamond · 80.5 source-defined score; unit unspecified
mmlu_pro · 23.89 source-defined score; unit unspecified
pass-rate-strict · 9.18 source-defined score; unit unspecified
diamond · 82.7 source-defined score; unit unspecified
lexam_hard · 36.18 source-defined score; unit unspecified
id · 83.9 source-defined score; unit unspecified
warping · 87.77 source-defined score; unit unspecified
photographed · 78.4 source-defined score; unit unspecified
layout · 63.37 source-defined score; unit unspecified
mmlu_pro · 45.95 source-defined score; unit unspecified
mmmu_pro_vision · 75.3 source-defined score; unit unspecified
scanning · 89.67 source-defined score; unit unspecified
zh · 46.6 source-defined score; unit unspecified
overall · 83.65 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.