dsh-model-switch
DSH plugin: choose models for subagents and plan execution separately / 为子代理和计划执行选择更合适的模型
20 results
DSH plugin: choose models for subagents and plan execution separately / 为子代理和计划执行选择更合适的模型
DSH 插件:给所有非官方(自定义)提供商的模型自动填充模型信息,包括:推理级别、最大上下文、输出上限与图片模态,数据来自 models.dev。
DSH 模型中心插件:免费模型目录 / 真免费实测 / 价格余额 / 一键切换 / 用量统计 — Model center plugin for DeepSeek Harness
DeepSeek Harness 插件:按工作阶段自动路由模型(对话/规划用 flash,实施用 pro)。
Show a hover hint with a model's full configuration (input types, context window, max tokens, protocol, base URL, credential env, …) over each model in the DeepSeek Harness web model picker, read from ~/.dsh/settings.yaml.
Type-to-search filtering for DeepSeek Harness model selection lists
DeepSeek Harness plugin: LLM model selection & deployment analysis (deployability, VRAM, TTFT, latency, throughput, power for 38 models × 20 GPUs/NPUs)
为 DeepSeek Harness 提供持久化模型别名、别名选择器和设置页面
Configurable model-request retry limit for DeepSeek Harness (Settings → General).
众数
A DeepSeek Harness plugin with a vertical model list and an animated Three.js particle reasoning-effort control.
mycel working modes: a switch_mode tool returning one selected mindset as an ordinary tool result, plus a shared core-behavior system-prompt section
Model availability tester panel in the Models settings page: one-click real streaming test per added model with TPS / first-token / elapsed metrics
在 dsh 模型目录的每一行直接设置该模型的思考强度范围与支持的输入模态 (Per-model reasoning-effort range and accepted input modalities, right in the dsh Models catalog)
自定义模型/提供方顺序:模型设置页里拖拽排序,宿主的模型列表按这份顺序返回(写回配置真源,备份去重且只留最近 N 份)
Expose per-model thinking levels (reasoningEfforts) inside the Models settings card for a custom model API route.
Route selected models through an outbound proxy in DeepSeek Harness, falling back to a direct connection while that proxy is unavailable.
Run the same prompt through every model you have, and see the difference without squinting.
Model Check:DeepSeek Harness 桌面 / Web 通用的模型检测插件,支持 ModelTrace 与糖果题联合检测、响应路由及行为指纹。
Per-model reasoning effort levels in Settings → Models: a levels editor beside context window and input types that converts each selection into the model's reasoningEfforts in the profile's cordis.patch.yml.