dsh-plugin-rollout-scout
Detect which conversation model your DeepSeek Harness account is served: score probe conversations' live chain-of-thought, cancel old-model probes in seconds, keep confident catches.
17 results
Detect which conversation model your DeepSeek Harness account is served: score probe conversations' live chain-of-thought, cancel old-model probes in seconds, keep confident catches.
灰度模型检测工具:一键开测 40 个并发 dsh 会话(V4 Pro + max 思考),流式监听思维链——出现 I'm/I'll 特征标记为灰度神模型,出现 Let me 标记为正常模型;统计灰度出现率,结果可在设置页查看。
DeepSeek balance banner for DSH: top-right USD/CNY balance chips with damage/gain animations and peak/valley detection
Real-time repetition guard for DeepSeek Harness (DSH): stops model generation when the same string repeats >=10 times in the streamed output. 实时检测 DSH 大模型流式输出中的重复内容,同一字符串重复十次以上立即停止生成。
Local repository intelligence for DeepSeek Harness: stack detection, dependency maps, and module-reference analysis 鈥?no extra services, everything runs on your filesystem.
输入一张图像,对其中的人头人体进行检测,输出图片中所有目标的检测框、置信度和标签。
输入一张图像,对其中的电动自行车进行检测,输出图片中所有目标的检测框、置信度和标签。
输入一张图像,对其中的手机进行检测,输出图片中所有目标的检测框、置信度和标签。
Result-aware stuck-loop guard for DeepSeek Harness: advisory nudges plus a monotonic hard stop. Only repeats with identical results count, so progressing polls and post-edit re-reads never trigger.
输入一张图像,输出图像中所有识别到的野生动物的检测框、置信度及标签。
电商素材工坊(中英双语)。用户需要生成电商主图、详情图、场景图、产品图合成、品牌叠加时使用。自动识别品类→匹配风格→场景感知合成→统一文字→自动质检→多平台适配→标准化交付。E-commerce product image studio: category detection, style matching, scene-aware compositing, text overlay, quality check, multi-platform adaptation, batch delivery.
DeepSeek Harness plugin conflict checker: pre-flight duplicate-id detection and optional real startup smoke test.
CalDAV + iCalendar + RRULE calendar integration bundle for DeepSeek Harness (dsh), with Chinese-bias (lunar calendar / holidays / Asia/Shanghai) and conflict detection.
检测各类通用场景中出现的火焰,最佳使用场景:安防摄像头、交通摄像头视角。
输入一张图像,对其中的人/宠物/车/火焰/纸箱进行检测,输出图片中所有目标的检测框、置信度和标签。
输入一张图像,对其中的手势进行检测,输出图片中所有目标的检测框、置信度和标签。
输入一张图像,对其中的香烟目标进行检测,输出图片中所有目标的检测框、置信度和标签。