@xidong-ai/dsh-rate-limiter

DeepSeek Harness 主动限速插件：按供应商令牌桶限速，超限延迟排队而非失败适合需要防止 API 调用超限、通过延迟排队平滑请求的用户。

パッケージ

@xidong-ai/dsh-rate-limiter

互換性

未検証

Harness ピア範囲

^0.1.0-rc.7 || ^0.1.1-rc.1

バージョン

0.3.2

ライセンス

MIT

最終更新

2026/09/12

Configuration

Configure the token bucket per provider in the profile's cordis.patch.yml (or this plugin's cordis.patch.yml):

- id: rate-limiter
  config:
    enabled: true
    providers:
      nvidia:
        rate: 0.5        # tokens/second (long-term average QPS)
        burst: 1         # bucket capacity (allowed burst requests)
      sensenova:
        rate: 0.02778
        burst: 1

rate: refill rate (tokens/second), i.e. the long-term average request rate.
burst: bucket capacity, the number of burst requests allowed.
Providers not listed are not rate-limited; requests pass through untouched (zero intrusion).
enabled: false disables the plugin entirely.

@xidong-ai/dsh-rate-limiter

インストール

ドキュメント

Configuration