221 results
DeepSeek Harness client plugin: adds an 'Invert selection' button to the fetch-models dialog of the Models settings section, so only the rows that are already in the model list stay checked on a second fetch + add pass.
Diagnose WSL/Windows proxy, Node 24 fetch, dsh web process env, proxy TCP listen, DeepSeek/npm/ModelScope reachability.
DeepSeek Harness LLM adapter plugin for locally deployed Qwen models behind a vLLM OpenAI-compatible endpoint: per-model multimodal switch, fully configurable reasoning efforts, and a web settings page (client plugin) for editing the deployment from the frontend
Model capability radar for the DeepSeek Harness web GUI: a Settings tab that charts codexradar crowd-benchmark scores per model tier — task-composition bars, 7-day IQ trend, efficiency badges
Give text-only models eyes: an analyze_image tool for DeepSeek Harness, backed by free Chinese vision APIs (GLM-4V-Flash / Qwen-VL) or any OpenAI-compatible vision endpoint. 给纯文本模型装上眼睛。
Universal local vision bridge tool for DSH: describe/answer questions about images via any OpenAI-compatible local inference framework (llama.cpp / Ollama / LM Studio / vLLM), vision models auto-detected
deepseek-eyes：给 DeepSeek Harness 的纯文本模型装上眼睛——把粘贴的图片经 OpenAI 兼容视觉 API 转成文字描述，再转发给真实的文本模型（Give text-only models in DeepSeek Harness eyes: describe pasted images via an OpenAI-compatible vision API, then forward to the real provider. Not affiliated with DeepSeek.）
DSH-native vision bridge for text-only models with native image attachments, multi-image evidence batching, and session-scoped validated Evidence caching.
AI 模型
Use OpenCode models inside DeepSeek Harness without a separate provider API key.
Compare multiple coding models side by side in DeepSeek Harness with isolated Git worktrees
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。
Sync CommandCode plan models (goat/pro/max) from the official models API + GOAT plan catalog into llm-pi-ai providers commandcode-<plan>-autosync (one-click create/update in settings).
dsh-eye-vision: give text-only DeepSeek Harness models eyes — image understanding via any OpenAI-compatible multimodal API. Fork of dsh-free-vision with custom-provider fixes and an allowed-directories whitelist.
OpenRouter 免费模型面板：自动拉取实时免费模型清单（综合能力 1-5★ 分级、最优优先排序），一键安装进 openrouter 路由并切换当前对话模型（全局默认 + 会话级双写，推理档自动选该模型支持的最高档），免手动配置。
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.
Enhanced DSH subagent tools with per-call model and reasoning-effort selection
DSH web plugin: splits the composer model selector into two independent controls — a searchable, favorite-marking model dropdown and a reasoning-effort slider — plus Ctrl+P (cycle favorite models) and Ctrl+T (cycle reasoning effort) shortcuts.
Use ZenMux models in DeepSeek Harness with OAuth login, model search, and protocol-aware routing
Image routing for text-only models in DeepSeek Harness: a global analyze_image tool (Kimi vision) plus automatic rewriting of pasted images into attachment references when the active model cannot see images.
Grok (xAI) subscription OAuth provider for DeepSeek Harness: one-click login in Settings → Models, token refresh, live catalog, streaming over cli-chat-proxy.grok.com
Register AllRouter as a DSH provider: one base URL and one key for Kimi K3, Claude, GPT, Grok, GLM, DeepSeek, MiniMax-H3. Switch models without managing separate providers; every call reports real cost.
Automatic model routing for DeepSeek Harness: a virtual `smart` model classifies every request by difficulty (hard / normal / easy) and by vision need, then delegates to the models you already configured in Settings → Models. 内置智能路由：三级难度 + 视觉自动路由，模型从「设置-模型」已配置模型中选取，虚拟 smart 模型零配置接入。
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。