DeepSeek Harness Plugin Hub

Publish and manage complete Harness Profiles. Discover Plugins for your next setup.

Explore

PluginsPresetsDocsNews

Community

Publish a pluginContactReport an issue

Resources

Plugin Hub on GitHubDeepSeek HarnessSystem statusPrivacy notice
© 2026 DeepSeek Harness Plugin HubPowered byPaxTech

Independent and unofficial. Not affiliated with, authorized by, or endorsed by DeepSeek.

DeepSeek Harness Plugin Hub
ProfilesPluginsCategoriesNewsDocsSign inManage Profiles
ProfilesPluginsCategoriesNewsDocsSign in
← Plugins

@morlay/dsh-llm-openai-compatible

Llm Openai Compatible

OpenAI-compatible LLM adapter plugin for DeepSeek Harness with configurable default sampling parameters (temperature / topP / topK / penalties / seed) over a providers dict.

The plugin will be installed here. Keep web if you are unsure.

npx -y @deepseek-ai/dsh plugin --profile web add @morlay/dsh-llm-openai-compatible@0.0.9-alpha.8
READMECompatibilityVersions

Compatibility and provenance

Llm Openai Compatible is published as @morlay/dsh-llm-openai-compatible and currently resolves to version 0.0.9-alpha.8. The Hub verifies its manifest and preserves the exact installation source for reproducible installs.

DSH compatibility
*
Runtime surfaces
any
Release source
npm
Registry updated
11/4/2026

Versions

0.0.9-alpha.8prerelease
9/24/2026
0.0.8stable
9/14/2026
0.0.7stable
9/13/2026
0.0.6stable
9/11/2026
Show 17 more versionsCollapse versions
0.0.9-alpha.7prerelease
9/23/2026
0.0.9-alpha.6prerelease
9/22/2026
0.0.9-alpha.5prerelease
9/20/2026
0.0.9-alpha.4prerelease
9/20/2026
0.0.9-alpha.3prerelease
9/18/2026
0.0.9-alpha.2prerelease
9/18/2026
0.0.9-alpha.1prerelease
9/18/2026
0.0.9-alpha.0prerelease
9/18/2026
0.0.5stable
9/10/2026
0.0.4stable
9/10/2026
0.0.4-alpha.4prerelease
9/10/2026
0.0.4-alpha.2prerelease
9/8/2026
0.0.4-alpha.1prerelease
9/8/2026
0.0.4-alpha.0prerelease
9/8/2026
0.0.3stable
9/4/2026
0.0.2stable
9/3/2026
0.0.1stable
8/20/2026

Related plugins

Loading related plugins…

Latest
0.0.9-alpha.8
DSH
*
HMR
Process restart
Tree shaking
Safe tree shaking not declared
Unpacked size
105.6 kB
Files
20
Surface
any
License
MIT
Source
npm
GitHub
★ 9
Weekly downloads
343
Security scan
✓ v0.0.9-alpha.8 scan passed
Last push
11/4/2026
View source ↗
README badge

Click the badge to copy Markdown for your README.

Do you maintain this Plugin?Claim benefit · Priority security scan

Verify the GitHub repository declared in package.json to manage this listing. After you claim it, Hub will prioritize a security scan of the current version and publish the result when it passes.

Claim this Plugin →
Report an issue

Related plugins

More verified plugins in models-usage.

Usage@linxin666/dsh-usageUsage statistics plugin for the dsh web GUI: per-provider balance and coding-plan quota detection plus a live token usage ledger, with a compact price glance card above the sidebar's Settings rowConnect Workbuddydsh-connect-workbuddyConnects locally signed-in WorkBuddy models (domestic and international versions) to DeepSeek Harness, automatically reading each model’s points multiplier from the upstream catalog on every refresh, with per-model image input toggles, multi-account switching, automatic account rotation from an accoCost Meterdsh-cost-meterDeepSeek Harness session cost statistics plugin: tracks current session costs, today’s expenses, and historical records, with synchronization to official pricing. Supports billing across multiple vendors and models, with a built-in catalog of prices for 170+ model IDs and automatic matching. Also suClient Ui Model Capabilities@linxin666/dsh-client-ui-model-capabilitiesPer-model capability declarations (image input and reasoning efforts) for custom pi-ai providers, edited in place on the Models settings cards; writes the official llm-pi-ai settings namespace, no DSH source changes.

README

@morlay/dsh-llm-openai-compatible

DeepSeek Harness 的 OpenAI 兼容 LLM 适配器插件。与内置 llm-pi-ai / llm-deepseek 不同，本插件的配置 schema 支持 profile 级默认采样参数 （temperature / topP / topK / presencePenalty / frequencyPenalty / seed），请求级 GenerateOptions.temperature 优先于 profile 默认值；providers 用 dict 多路由结构（与 llm-pi-ai 一致）。

传输层复用 @ai-sdk/openai-compatible （wire 序列化与 SSE 解析由 SDK 负责）；本插件负责 harness 消息 → AI SDK prompt 转换、采样默认合并、stream part → StreamChunk 翻译、错误归一化与凭据策略。

本文件只给配置面与用法；每条规则与取舍的 home 是各自 ADR（见下）。

配置

providers 是 dict：key 就是 provider 路由键（选择器与 GenerateOptions.provider 使用），值是 profile。它标了 .volatile()——运行期可改： 值经 Loader 的引用读取，设置页（Models 页）改的就是它，不需要重挂这行插件（见 ADR-运行期改配置走volatile引用）。

配置写在这行的 config 里（bundle patch / profile patch / 设置页都落到这里）：

- id: llm-openai-compatible
  name: "@morlay/dsh-llm-openai-compatible"
  config:
    providers:
      ollama:
        apiKeyEnv: OLLAMA_API_KEY
        baseURL: https://ollama.com/v1
        displayName: Ollama Gateway
        # === 采样默认参数（请求级 temperature 优先）===
        temperature: 1 # 0..2
        topP: 0.95 # 0..1 → wire top_p
        topK: 40 # 正整数 → wire top_k（非标准，仅网关支持时发送）
        presencePenalty: 0 # -2..2 → wire presence_penalty
        frequencyPenalty: 0 # -2..2 → wire frequency_penalty
        seed: 42 # 正整数 → wire seed
        # === 推理 ===
        reasoning: high # 部署默认档位（省略 = 提供方默认）
        # === 模型目录 ===
        defaultContextWindow: 262144
        defaultMaxTokens: 32768
        models:
          - id: deepseek-v4-flash:0731
            name: DeepSeek V4 Flash
            contextWindow: 1000000
            maxTokens: 65535
            inputModalities: [text, image]
            reasoningEfforts:
              off: # off 空值 = 不发送 reasoning_effort
              high: high # 档位 → wire reasoning_effort 拼写
              max: max
        # === 传输 ===
        maxRequestImageBytes: 20971520
        streamIdleTimeoutMs: 300000
        timeoutMs: 600000 # 整体请求超时；缺省不设
        retryPolicy:
          mode: normal
          maxRetries: 5

旧 $DSH_HOME/settings.yaml 的 llm-openai-compatible 段由上游启动时一次性导入到同 id 的行 config。

规则与取舍

规则细节（合并表、wire 字段落点、端点与错误映射）在各自 ADR 里维护，这里只列结论：

  • 采样默认值与省略语义：请求级优先于 profile 默认，缺省一律不发送（= 提供方 默认）——见 ADR-采样默认值合并规则与省略语义。
  • 模型目录缺省为空、描述模型绝不抛错：未列出的 id 原样透传，不支持的能力配置推迟到 请求执行处失败——见 ADR-模型目录缺省为空且描述模型绝不抛错。
  • 传输层复用 SDK：非标准字段（top_k）与用量方言在本层显式处理——见 ADR-传输层复用ai-sdk-openai-compatible而非自研wire序列化。
  • 凭据经 ctx.credentials 解析（服务缺失回退 launch environment）：见 ADR-凭据经credentials服务解析而非直接读环境变量。
  • 图片超预算报错交由 durable offload 重试（本适配器不自行裁剪请求图片）：见 ADR-图片超预算报错交由durable-offload重试。
  • 多路由结构对齐 llm-pi-ai（数组式 profiles 被拒绝）：见 ADR-providers采用dict多路由结构对齐llm-pi-ai 与 ADR-起因llm-pi-ai的请求参数配置不完整。
  • 未做（YAGNI）：modelOverrides、模型 discovery（GET /models）、OAuth / 非 bearer 认证——理由见模型目录 ADR。

从 llm-pi-ai 迁移

把 llm-pi-ai.providers.<route> 的 baseURL / models / 采样字段平移到 llm-openai-compatible.providers.<route>（无 api 字段——协议固定 chat-completions），apiKeyEnv 与 retryPolicy 原样保留；reasoningEfforts 的 off 空值语义一致。

构建、测试与类型检查的命令见根 justfile 与 mise.toml。