dsh-token-billing
Run the following command in DeepSeek Harness:
dsh plugin install jianzhexu/dsh-token-billing
Paste the following prompt into your AI chat to install this plugin:
Run "dsh plugin install jianzhexu/dsh-token-billing" in the DeepSeek Harness terminal to install; the source is at https://github.com/jianzhexu/dsh-token-billing and the plugin mounts into the Web UI stats band automatically.
About this plugin
Running inference on DeepSeek can quietly rack up dozens of model calls per session. Add model switches and workspace hops, and the real cost becomes a blind spot until the monthly bill arrives. dsh-token-billing closes that gap by logging every call at DeepSeek official pricing in real time, turning cost tracking from a retrospective exercise into a glanceable live metric.
The core experience is a persistent billing readout row under the Web UI input box. It names the active workspace inline and updates totals as you switch; hovering reveals a full breakdown covering per-model token usage, session and global cost, and a built-in peak and off-peak price table. Pricing logic is bundled with the plugin and refreshed per release, so unlisted models are simply not billed rather than guessed. Ledger data persists in a dedicated storage domain and survives restarts.
It is aimed at developers who call DeepSeek APIs at high frequency, teams juggling inference costs across multiple workspaces, and anyone who wants token spend to be a first-class, visible part of their daily workflow. The plugin is plain JavaScript with no build step, so it mounts cleanly on install and can be removed with a single line in your profile config.
Use Cases
- Monitor live token spend per workspace and globally during a session
- Calculate true cost of an inference task under peak and off-peak pricing
- Track cumulative fees while switching workspaces and hover for a full per-model breakdown
Best For
- Developers making high-frequency DeepSeek API calls
- Teams managing inference costs across multiple workspaces
- Anyone who wants token spend as a first-class daily metric
Related Plugins
A unified suite combining hot runtime injection, task-aware thinking-mode routing, and a graded session protocol with red-team gates to sustain model diligence across long-horizon inference.
ModLens is a vision plugin for DeepSeek Harness that gives text-only models sight by reading images pasted directly into chat, with zero-config setup and multiple vision engines.
On-demand vision for text-only DeepSeek Harness agents: built-in free keyless vision chain and 14 vision tools, routing image turns as tool calls to vision models with pixel fidelity, no Python needed, one-command install.
Give text-only models in DeepSeek Harness eyes, enabling image Q&A, long-screenshot OCR, UI restoration, and GUI visual tasks.