Usage & Billing

Cost tracking for DeepSeek Harness: session and model costs, token usage, budgets, provider balances, and coding-plan quotas. Bilingual English/Chinese UI.

Provider balances, subscription quotas, and token-usage analytics for the DeepSeek Harness Web GUI (dsh web).

DeepSeek account balance and session cost readout for the DeepSeek Harness Web GUI

Langfuse observability for DeepSeek Harness (dsh): exports agent sessions as OpenTelemetry trace trees (GenAI semconv) to Langfuse's OTLP endpoint

按请求粒度记录模型 token 用量（按天 JSONL，首次启动自动补齐历史），Web 设置面板「Token 用量」统计页（GET /token-usage/stats 路由）

RMB/USD token-billing for dsh web: official-policy auto pricing with peak/off-peak hours, per-message ledger, account balance polling, locale-driven currency display; host /billing route + hand-written client fee badge/panel (read-only over session/event).

Persistent token usage records and dashboard for DeepSeek Harness

A right-side floating panel and /context command for DeepSeek Harness — a live donut chart of context token usage, allocation, and estimated cost.

Token 数据面板：历史累计、趋势图、活跃热力图、模型用量分析、CSV/JSON 导出；不保存提示词/回复正文/API key

A DeepSeek Harness (dsh) bundle that shows OpenCode Go subscription usage in the Web GUI's composer dock — the same seat as the built-in conversation stats line.

Cost governance for DeepSeek Harness: aggregated token/cost metering per model, session and day, budget caps with threshold alerts and over-limit policies, carbon footprint estimation, per-model latency benchmarks, a Settings budget tab, and the /budget command

Multi-provider AI balance, quota, and token usage for the dsh sidebar, with a daily heatmap.

用量与消费统计

What a DeepSeek Harness run costs in tokens, money, and time. OpenTelemetry traces, metrics, and logs in GreptimeDB, with seven Grafana dashboards.

Token usage & cost monitor for DeepSeek Harness — floating widget with multi-dimensional stats, time-series charts, auto-detected billing plans (Code/Token) and estimated spend.

DeepSeek Harness billing plugin: account balance + per-model remaining-task estimator with a session-header badge

Low-cost video understanding tool: a video_understand tool turns a Bilibili link / BV / local video into an AVIS info layer (ASR + scene structure + object tracks + YOLO labels) and returns summary + Q&A. Question-driven dynamic routing across layers (L0 ASR / L1 object tracks / L2 key-frame VLM), semantic-layer reuse for repeat questions, per-question budget cap. Python engine: core layer needs faster-whisper / opencv / yt-dlp (~200-300MB); the optional semantic layer adds ~2GB of torch / transformers / ultralytics. A bundled doctor --fix sets up the venv and installs both.

Analyze usage across models, providers, workspaces, and time ranges — including custom start/end dates over the full retained history — with live cache-hit visibility and separate model and provider breakdowns. Includes hourly single-day trends, daily cross-day trends, a 53-week GitHub-style heatmap, usage streaks, model and workspace Token-share donut charts with Token, cost, and percentage details, independent combinable filters, cache-efficiency trends, DeepSeek account balance, workspace aliases, CSV export, and a durable ledger that preserves successfully flushed usage after session deletion — and after a workspace is deleted, where its recorded usage is summed into one Deleted row — and rebuilds incrementally — on restart, unchanged sessions are applied from the ledger without re-reading their logs (a per-session log revision is the change signal), and only changed/new events are folded. The workspace registry is kept fresh automatically through the DSH domain/changed probe: only added/removed workspaces are reprocessed incrementally, unchanged workspaces reuse their computed aggregates and ledger with zero rescan, and usage is strictly limited to DSH-registered workspaces (unregistered cwds are ignored). After a scan, lightweight revision status fetches full history only when the Host/data state changes and exposes non-sensitive sync health (revision reuse, rereads, ledger recovery, and failures). Token accounting is explicit: input is fresh (cache read/write and reasoning are separate buckets) and zero-usage replays never overwrite recorded usage. Chinese uses local time, English uses UTC; API access is loopback-only.

Header quota badge plus a usage-monitoring tab: live balance/quota remaining and reset countdown per provider, with requests, tokens (input/output/cache-read/cache-write), cache hit rate, estimated cost and average latency tracked per session, provider, model and client, stored locally in SQLite with cross-device import/export.

During Beijing working-day peak hours (default 09:00-12:00 and 14:00-18:00) it blocks official DeepSeek requests or reroutes them to a configured target provider; off-peak traffic flows to official as usual.

Developer tool: live token usage & cost monitoring for DeepSeek Harness - consumed tokens for the current session and across all sessions, read from token-meter projections. No model calls.

LLM quota monitor plugin for the DeepSeek Harness web GUI: breathing indicator, rolling / weekly / monthly usage panel · by Xu Yuanshan

A configurable 17-segment session status bar for the dsh composer dock with live TPS, per-model cost estimation, and a usage & cost dialog.

Configurable DSH usage statistics: a sidebar text/card widget plus a detail dialog with per-provider and per-model breakdowns, cache-hit rates, token trends, and latency details (TTFT and duration).