dsh-subagent-model-router
Model-selectable subagents and guided model routing setup for DeepSeek Harness
18 results
Model-selectable subagents and guided model routing setup for DeepSeek Harness
Thin routing policy for DeepSeek Harness: keyword routing, allowlisted model switching, and isolated image generation. / DeepSeek Harness 薄路由策略:关键词路由、白名单模型切换和隔离生图。
DeepSeek Harness 峰谷定价自动模型路由插件:高峰时段自动使用 deepseek-v4-flash,空闲时段自动使用 deepseek-v4-pro;用户显式选择的模型始终优先。
Unified ModelID routing for DeepSeek Harness: one logical model id over multiple providers, first-token failover with cooldown, health-aware ranking, fallback reasoning efforts, purpose-based simple/complex task split, per-tier reasoning effort, and a settings-page management panel.
DSH 插件:按任务复杂度在 deepseek-v4-flash 与 deepseek-v4-pro 之间自动路由(只升不降)
Hybrid model router for DeepSeek Harness: declarative rules, user-configurable cost control (cost-first / quality-first / balanced), optional LLM task classifier, and failure fallback chains.
Native DSH provider hub with named API keys, per-key model allowlists, OAuth accounts, failover, and redacted logs
Auto-switch DeepSeek models for DeepSeek Harness: routes every model request to the vision model when images are present, to the pro model for complex tasks, and to the fast model otherwise; includes a switch_model tool for manual per-session overrides.
Budget-LLM pool router for DeepSeek Harness: usage ledger, health tracking, Thompson-sampling adaptive failover across free-tier models, quality tier ladder, executor pool, paid-guard, web panel + reroute badge. 穷鬼路由器。
Flash-only reasoning-effort routing for DeepSeek Harness: classifies each step by its shape and sets the DeepSeek flash model's reasoning effort to match. The model never changes; effort max is opt-in.
Policy-only multi-source quota router for DeepSeek Harness: deterministic task profiles, ordered candidate chains, health-aware fallback, and observable decisions.
Configurable task-aware model routing bundle for DeepSeek Harness
DSH plugin: a virtual model-router provider that aggregates every configured provider and model, routes each request to the model's currently active provider, and ships a settings page for per-model provider ordering and switching.
Deterministic per-turn adaptive model routing for DeepSeek Harness
智能体选择器:像模型选择器一样在输入框旁勾选委派目标+具体模型,把对话任务派给外部智能体(Codex CLI / WorkBuddy 内置模型活动价 automation 桥 / WorkBuddy 自定义模型直连 / Claude Code)执行,结果流回 DSH 对话。含 agent_dispatch 工具 + composer 两步下拉 + 设置面板 + 桥健康探活 + 反冒充出处凭证。
DeepSeek Harness plugin: route an agent request to a vision model (deepseek-v4-flash-vision-exp) when the user message contains an image, and keep the original model otherwise.
DSH plugin: declare VIRTUAL model ids bound to a pluggable routing algorithm (priority, round-robin) over real provider/model candidates; using the virtual id transparently dispatches to a real provider chosen by the algorithm, with failover.
DSH Effort Router(模型分流):按每一轮请求的难度自动选择模型与思考强度——简单问题走便宜快模型,难题才叫强模型。规则判定零 token,灰区交给一次小模型裁定,图片档强度随难度升降(low/high/max)。每轮请求级覆盖,不改你的会话选择与默认模型。