Evidence coverage 3498 published observations across 337 measures. Capability scoring groups compatible evidence into 44 benchmark families.
A published result contributes only when a matching reference-panel result is available. Missing evidence stays unknown. Coverage describes breadth, separately from performance.
Labs Models Families Evidence matrix Coverage across our labs Models and contributing families Raw published counts include all source types. Contributing and independent family counts follow the selected evidence policy.
16 models match “DeepSeek”. Clear search
Family coverage Benchmark family Capability Models contributing With independent support OSWorld Agentic 21 6 Webarena Verified Agentic 4 0 Browsecomp Agentic 22 0 Toolathlon Agentic 15 0 MCP-Atlas Agentic 19 0 AutomationBench Agentic 21 0 DeepSearchQA Agentic 10 0 GDPval Agentic 64 45 Bfcl V4 Agentic 12 0 Apex Agents Agentic 10 0 OfficeQA Agentic 11 0 GDP.pdf Agentic 14 0 Terminal-Bench Science Agentic 7 0 Humanity’s Last Exam Hard reasoning 32 9 GPQA Hard reasoning 45 18 ARC-AGI Hard reasoning 60 60 Aime Hard reasoning 22 0 FrontierMath Hard reasoning 10 0 Imo Answerbench Hard reasoning 5 0 SWE-bench Verified Coding 38 19 SWE-bench Pro Coding 20 4 SWE-bench Multilingual Coding 9 0 DeepSWE Coding 33 28 LiveCodeBench Coding 25 9 Terminal-Bench Coding 22 22 Scicode Coding 0 0 Human preference (Arena) Human pref 119 119 MMLU-Pro Knowledge 68 52 Gmmlu Knowledge 3 0 Milu Knowledge 3 0 Simpleqa Verified Knowledge 2 0 Mmmu Multimodal 2 0 Mmmu Pro Multimodal 24 0 CharXiv Multimodal 16 0 Mathvista Multimodal 2 0 Chartography Multimodal 3 0 Lvbench Multimodal 6 0 ScreenSpot Multimodal 12 0 Ocrbench V2 Average Accuracy Multimodal 12 0 MRCR Long context 12 0 Longbench Long context 8 0 Ruler 1M Long context 2 0 Graphwalks Long context 7 0 Aa Lcr Long context 5 0
Model evidence matrix Showing the same 1–16 models as the list above.
Model OSWorld Webarena Verified Browsecomp Toolathlon MCP-Atlas AutomationBench DeepSearchQA GDPval Bfcl V4 Apex Agents OfficeQA GDP.pdf Terminal-Bench Science Humanity’s Last Exam GPQA ARC-AGI Aime FrontierMath Imo Answerbench SWE-bench Verified SWE-bench Pro SWE-bench Multilingual DeepSWE LiveCodeBench Terminal-Bench Scicode Human preference (Arena) MMLU-Pro Gmmlu Milu Simpleqa Verified Mmmu Mmmu Pro CharXiv Mathvista Chartography Lvbench ScreenSpot Ocrbench V2 Average Accuracy MRCR Longbench Ruler 1M Graphwalks Aa Lcr DeepSeek V4-Pro 0813 Unknown Unknown Unknown 50.0 Provider reports · 7/7 reference peers
Unknown 75.0 Provider reports · 7/7 reference peers
Unknown 76.2 Independent support · 42/42 reference peers
Unknown Unknown Unknown Unknown Unknown 70.8 Provider reports · 7/7 reference peers
100.0 Independent support · 2/2 reference peers
58.5 Independent support · 59/59 reference peers
Unknown Unknown Unknown Unknown Unknown Unknown 59.0 Independent support · 27/27 reference peers
Unknown Unknown Unknown 80.3 Independent support · 76/76 reference peers
Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown DeepSeek-V4-Flash-0731 Unknown Unknown Unknown 25.0 Provider reports · 4/4 reference peers
Unknown 25.0 Provider reports · 4/4 reference peers
Unknown Unknown Unknown Unknown Unknown Unknown Unknown 0.0 Provider reports · 4/4 reference peers
Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown 26.4 Independent support · 27/27 reference peers
Unknown Unknown Unknown 60.5 Independent support · 76/76 reference peers
Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown DeepSeek-R1-0528 Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown 75.0 Independent support · 8/8 reference peers
Unknown Unknown 78.0 Independent support · 41/41 reference peers
74.5 Independent support · 51/51 reference peers
Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown DeepSeek-V3.2 Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown 20.3 Independent support · 59/59 reference peers
Unknown Unknown Unknown 66.7 Independent support · 18/18 reference peers
Unknown Unknown Unknown Unknown Unknown Unknown 52.6 Independent support · 76/76 reference peers
Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown DeepSeek V4.1 Flash Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown 69.5 Independent support · 59/59 reference peers
Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown DeepSeek V4 Flash 0424 Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown DeepSeek V4-Pro 0424 Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown DeepSeek-R1-0528-Qwen3-8B Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown DeepSeek-R1-Distill-Llama-70B Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown DeepSeek-R1-Distill-Llama-8B Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown DeepSeek-R1-Distill-Qwen-1.5B Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown DeepSeek-R1-Distill-Qwen-14B Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown DeepSeek-R1-Distill-Qwen-32B Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown DeepSeek-R1-Distill-Qwen-7B Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown DeepSeek-V3.2-Speciale Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown DeepSeek-V4-Flash-Vision-Exp Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown
Family definitions and scoring rules · Compare capabilities · Compare published evidence