DeepSeek-V4-Flash (思考 · API)Normal level for large API models — the 284B parameter advantage is clear. Results on this page are for reference only.
DeepSeek-V4-Flash · OpenCode API · 作为参考reference
加权总分Weighted Total
94.0
加权原始分 = TC×0.3 + BF×0.3 + HA×0.4Weighted raw score = TC×0.3 + BF×0.3 + HA×0.4
实用得分Effective Score
85.0
−9 重试扣分−9 retry penalty
ToolCall-15
100
15/15 通过 · 100% 🏆15/15 passed · 100% 🏆15/15 passed · 100% 🏆 · 重试Retry -2
BugFind-15
93.3
14/15 通过 · 93%14/15 passed · 93% · 重试Retry -5
HermesAgent-20
90
16/20 通过 · 80%16/20 passed · 80% · 重试Retry -2
📊 计分规则Scoring Rules — 能力上限Max Score = 加权原始分 = ToolCall×0.3 + BugFind×0.3 + HermesAgent×0.4= Weighted raw score = ToolCall×0.3 + BugFind×0.3 + HermesAgent×0.4
实用得分Effective Score = 能力上限 − 重试扣分（每题首次通过不扣分，重试后通过每重试1次扣1分，重试后仍失败不参与扣分）。Max Score − retry penalty (first pass = no penalty, each retry = +1pt, failed retries excluded).
实用得分Effective Score = 能力上限 − 重试扣分（每题首次通过不扣分，重试后通过每重试1次扣1分，重试后仍失败不参与扣分）。Max Score − retry penalty (first pass = no penalty, each retry = +1pt, failed retries excluded).