DeepSeek V4 Flash Vision
公开成绩与计算依据
综合参考分根据已公开的评测资料整理。查看发布日期依据
综合参考分 62.4 · 10项基准
推理求解 74.9
GPQA Diamond · 0.913131313131313 比例
Artificial Analysis · 独立评测 · Diamond
DeepSeek V4 Flash Vision (Max) · Current AA standalone protocol. Status verified in aa-method-current.md section gpqa-diamond
HLE · 0.211 比例
Artificial Analysis · 独立评测 · Diamond
DeepSeek V4 Flash Vision (Max) · Artificial Analysis current published protocol; hle-diamond
工具与任务执行 68.7
tau3-Banking · 0.410309278350515 比例
Artificial Analysis · 独立评测 · 1.0.1
DeepSeek V4 Flash Vision (Max) · Current AA standalone protocol. Status verified in aa-method-current.md section tau3-banking
GDPval-AA · 1554.01 分
Artificial Analysis · 独立评测 · 2.1
DeepSeek V4 Flash Vision (Max) · Artificial Analysis current published protocol; gdpval-aa-normalized
Terminal-Bench · 0.121212121212121 比例
Artificial Analysis · 独立评测 · 4.0
DeepSeek V4 Flash Vision (Max) · Artificial Analysis current published protocol; terminalbench-4-0
AutomationBench · 0.4747087040091967 比例
Artificial Analysis · 独立评测 · current published task suite
DeepSeek V4 Flash Vision (Max) · Artificial Analysis current published protocol; automationbench-aa
AA-Briefcase · 1418.4 分
Artificial Analysis · 独立评测 · 1.1
DeepSeek V4 Flash Vision (Max) · Artificial Analysis current published protocol; aa-briefcase
EnterpriseOps-Gym · 0.4709042076991943 比例
Artificial Analysis · 独立评测 · current published task suite
DeepSeek V4 Flash Vision (Max) · Artificial Analysis current published protocol; enterprise-ops-gym-aa
语言与信息处理 55.3
GDP.pdf · 0.12 比例
Artificial Analysis · 独立评测 · current published task suite
DeepSeek V4 Flash Vision (Max) · Artificial Analysis current published protocol; gdp-pdf
多模态理解 29.4
MMMU-Pro · 0.747976878612717 比例
Artificial Analysis · 独立评测 · Standard 10
DeepSeek V4 Flash Vision (Max) · Artificial Analysis current published protocol; mmmu-pro