dsh-llm-cost
Per-turn, per-step LLM cost metering: a costUsage session projection, a cost line under each message, and an LLM+web price auto-maintenance tool.
每个 turn、每个 step 的 LLM 成本计量：costUsage 会话投影 + 消息下方成本行 + LLM 联网自动更新价格。
How to install
dsh plugin add github:chenyinrusi/dsh-llm-cost About
dsh-llm-cost 让 DeepSeek Harness 显示**每个 turn、每个 step 的 LLM 成本**（美元），并在价格过期时用 LLM + 联网自动更新价格表。 DSH 已经精确记录了每个 step 的真实 token 用量（assistant/message 的 usage：输入/输出/缓存读/缓存写），但**没有任何美元成本概念**。dsh-llm-cost 补上缺失的那一环：把「用量 × 单价」变成 DSH 的持久事实（costUsage session projection），并在每个对话框下方的 stats 行里渲染出该 turn 的成本。 工作原理 **host 端（node）**：一个 session projection 单元（照 token-meter 的骨架），在 session/event 上折叠： request/context（解析后的 provider/model 路由，post-fallback）决定每个 step 用的模型； assistant/message（或 assistant/chunk 的 usage chunk）报告该 step 的用量； 单价表 → costUsd，同一 turn/step 的重复采样做**替换**而非重复累计。 **client 端（浏览器）**：注册到 conversation.chat.tu…
Recommendation signals
Meta
- License
- MIT
- Language
- TypeScript
- GitHub stars
- 0
- mo. downloads
- –
- Last push
- 2026-08-16
- Created
- 2026-08-16
Links
Basic safety check
- Findings
- curated 收录但无 npm 包/安装命令
- Sources
- curated:awesome-dsh-plugin.com, curated:awesome-dsh-plugin/awesome-dsh-plugin
- Topics
- deepseek-harness, dsh-plugin, llm-cost
Related plugins
dsh-balance-meter
Ghost011118/dsh-balance-meter
DeepSeek account balance and session cost in the composer dock, with auto-fetched official pricing and peak/off-peak support.
TokenLedger
zh667/TokenLedger
Sidebar usage panel that attributes tokens to the relay site that served each request, read from your existing provider config: today/month/all-time totals, per-site and per-model breakdowns, a year activity heatmap, and New API / Sub2API / DeepSeek balances.
dsh-usage-billing
940842546/dsh-usage-billing
Usage and cost statistics with peak/off-peak pricing after the 2026-08-17 rate change, a floating summary panel, per-session breakdowns, and a day/week/month/year/all usage heatmap.
dsh-token-usage
LeemanCheung/dsh-token-usage
Local-first four-bucket Token observability: persistent session/provider/model/day dashboards, trends, budget and anomaly signals, public-rate estimates, safe aggregate exports, and explicit opt-in usage or session-trajectory analysis.