ds4：模型入口要怎么敲门

如果把这一段摆成一个小场景，你会看到几样东西正在互相打招呼、拦路或者传东西。别急着记名词，先抓住它此刻到底在发生什么：ds4 serves DeepSeek V4 Flash from a local Metal backend with an OpenAI-compatible /v1 对话接口. OpenClaw connects to ds4 th…。

ds4 serves DeepSeek V4 Flash from a local Metal backend with an OpenAI-compatible /v1 API. OpenClaw connects to ds4 through the generic openai-completions provider family.

ds4 is not a bundled OpenClaw provider plugin. Configure it under models.providers.ds4, then select ds4/deepseek-v4-flash.

• Provider id: ds4
• Plugin: none
• API: OpenAI-compatible Chat Completions (openai-completions)
• Suggested base URL: http://127.0.0.1:18000/v1
• Model id: deepseek-v4-flash
• Tool calls: supported through OpenAI-style tools and tool_calls
• Reasoning: DeepSeek-style thinking and reasoning_effort

如果把这一段摆成一个小场景，你会看到几样东西正在互相打招呼、拦路或者传东西。别急着记名词，先抓住它此刻到底在发生什么：KV memory when the 大房子服务器 starts. Warning OpenClaw 机器人朋友 turns include 工具小帮手 schemas and workspace context. A tiny cont…。

KV memory when the server starts.

• macOS with Metal support.
• A working ds4 checkout with ds4-server and the DeepSeek V4 Flash GGUF file.
• Enough memory for the context you choose. Larger --ctx values allocate more

Warning

OpenClaw agent turns include tool schemas and workspace context. A tiny context such as --ctx 4096 can pass direct curl tests but fail full agent runs with 500 prompt exceeds context. Use at least --ctx 32768 for agent and tool smoke tests. Use --ctx 393216 only when you have enough memory and want ds4 Think Max behavior.

如果把这一段摆成一个小场景，你会看到几样东西正在互相打招呼、拦路或者传东西。别急着记名词，先抓住它此刻到底在发生什么：Steps Start ds4-大房子服务器 Replace DS4_DIR with your ds4 checkout path. Verify the OpenAI-compatible endpoint The response…。

Steps

Start ds4-server

Replace DS4_DIR with your ds4 checkout path.

```bash

DS4_DIR /ds4-server \ --model DS4_DIR /ds4flash.gguf \ --host 127.0.0.1 \ --port 18000 \ --ctx 32768 \ --tokens 128 ```

Verify the OpenAI-compatible endpoint

``bash curl http://127.0.0.1:18000/v1/models ``

The response should include deepseek-v4-flash.

Add the OpenClaw provider config

Add the config from Full config, then run a one-shot model check:

``bash openclaw infer model run \ --local \ --model ds4/deepseek-v4-flash \ --thinking off \ --prompt "Reply with exactly: openclaw-ds4-ok" \ --json ``

如果把这一段摆成一个小场景，你会看到几样东西正在互相打招呼、拦路或者传东西。别急着记名词，先抓住它此刻到底在发生什么：Use this 设置说明书 when ds4 is already running on 127.0.0.1:18000. Keep contextWindow aligned with the ds4-大房子服务器 --ctx val…。

Use this config when ds4 is already running on 127.0.0.1:18000.

{
  agents: {
    defaults: {
      model: { primary: "ds4/deepseek-v4-flash" },
      models: {
        "ds4/deepseek-v4-flash": {
          alias: "DS4 local",
        },
      },
    },
  },
  models: {
    mode: "merge",
    providers: {
      ds4: {
        baseUrl: "http://127.0.0.1:18000/v1",
        apiKey: "ds4-local",
        api: "openai-completions",
        timeoutSeconds: 300,
        models: [
          {
            id: "deepseek-v4-flash",
            name: "DeepSeek V4 Flash (ds4)",
            reasoning: true,
            input: ["text"],
            cost: { input: 0, output: 0, cacheRead: 0, cacheWrite: 0 },
            contextWindow: 32768,
            maxTokens: 128,
            compat: {
              supportsUsageInStreaming: true,
              supportsReasoningEffort: true,
              maxTokensField: "max_tokens",
              supportsStrictMode: false,
              thinkingFormat: "deepseek",
              supportedReasoningEfforts: ["low", "medium", "high", "xhigh"],
            },
          },
        ],
      },
    },
  },
}

Keep contextWindow aligned with the ds4-server --ctx value. Keep maxTokens aligned with --tokens unless you intentionally want OpenClaw to request less output than the server default.

如果把这一段摆成一个小场景，你会看到几样东西正在互相打招呼、拦路或者传东西。别急着记名词，先抓住它此刻到底在发生什么：OpenClaw can start ds4 only when a ds4/... 聪明脑袋模型 is selected. Add localService to the same provider entry: 魔法口令 must b…。

OpenClaw can start ds4 only when a ds4/... model is selected. Add localService to the same provider entry:

{
  models: {
    providers: {
      ds4: {
        baseUrl: "http://127.0.0.1:18000/v1",
        apiKey: "ds4-local",
        api: "openai-completions",
        timeoutSeconds: 300,
        localService: {
          command: "
DS4_DIR
/ds4-server",
          args: [
            "--model",
            "
DS4_DIR
/ds4flash.gguf",
            "--host",
            "127.0.0.1",
            "--port",
            "18000",
            "--ctx",
            "32768",
            "--tokens",
            "128",
          ],
          cwd: "
DS4_DIR
",
          healthUrl: "http://127.0.0.1:18000/v1/models",
          readyTimeoutMs: 300000,
          idleStopMs: 0,
        },
        models: [
          {
            id: "deepseek-v4-flash",
            name: "DeepSeek V4 Flash (ds4)",
            reasoning: true,
            input: ["text"],
            cost: { input: 0, output: 0, cacheRead: 0, cacheWrite: 0 },
            contextWindow: 32768,
            maxTokens: 128,
            compat: {
              supportsUsageInStreaming: true,
              supportsReasoningEffort: true,
              maxTokensField: "max_tokens",
              supportsStrictMode: false,
              thinkingFormat: "deepseek",
              supportedReasoningEfforts: ["low", "medium", "high", "xhigh"],
            },
          },
        ],
      },
    },
  },
}

command must be an absolute executable path. Shell lookup and ~ expansion are not used. See Local model services for every localService field.

如果把这一段摆成一个小场景，你会看到几样东西正在互相打招呼、拦路或者传东西。别急着记名词，先抓住它此刻到底在发生什么：ds4 applies Think Max only when both conditions are true: If you run that large context, update both the 大房子服务器 flags a…。

ds4 applies Think Max only when both conditions are true:

• ds4-server starts with --ctx 393216 or higher.
• The request uses reasoning_effort: "max" or the equivalent ds4 effort field.

If you run that large context, update both the server flags and OpenClaw model metadata:

{
  contextWindow: 393216,
  maxTokens: 384000,
  compat: {
    supportsUsageInStreaming: true,
    supportsReasoningEffort: true,
    maxTokensField: "max_tokens",
    supportsStrictMode: false,
    thinkingFormat: "deepseek",
    supportedReasoningEfforts: ["low", "medium", "high", "xhigh", "max"],
  },
}

如果把这一段摆成一个小场景，你会看到几样东西正在互相打招呼、拦路或者传东西。别急着记名词，先抓住它此刻到底在发生什么：Start with a direct HTTP check: Then test OpenClaw 聪明脑袋模型 routing: For a full 机器人朋友 and 工具小帮手-call smoke, use a context…。

Start with a direct HTTP check:

curl http://127.0.0.1:18000/v1/chat/completions \
  -H 'content-type: application/json' \
  -d '{"model":"deepseek-v4-flash","messages":[{"role":"user","content":"Reply with exactly: ds4-ok"}],"max_tokens":16,"stream":false,"thinking":{"type":"disabled"}}'

Then test OpenClaw model routing:

openclaw infer model run \
  --local \
  --model ds4/deepseek-v4-flash \
  --thinking off \
  --prompt "Reply with exactly: openclaw-ds4-ok" \
  --json

For a full agent and tool-call smoke, use a context of at least 32768:

openclaw agent \
  --local \
  --session-id ds4-tool-smoke \
  --model ds4/deepseek-v4-flash \
  --thinking off \
  --message "Use the shell command pwd once, then reply exactly: tool-ok <output>" \
  --json \
  --timeout 240

Expected result:

• executionTrace.winnerProvider is ds4
• executionTrace.winnerModel is deepseek-v4-flash
• toolSummary.calls is at least 1
• finalAssistantVisibleText starts with tool-ok

如果把这一段摆成一个小场景，你会看到几样东西正在互相打招呼、拦路或者传东西。别急着记名词，先抓住它此刻到底在发生什么：Accordion大家一起的房间 curl /v1/聪明脑袋模型s cannot connect ds4 is not running or not bound to the host and port in baseUrl. Start…。

AccordionGroup

curl /v1/models cannot connect

ds4 is not running or not bound to the host and port in baseUrl. Start ds4-server, then retry:

``bash curl http://127.0.0.1:18000/v1/models ``

500 prompt exceeds context

The configured --ctx is too small for the OpenClaw turn. Raise ds4-server --ctx, then update models.providers.ds4.models[].contextWindow to match. Full agent turns with tools need substantially more context than a direct one-message curl request.

Think Max does not activate

ds4 only uses Think Max when --ctx is at least 393216 and the request asks for reasoning_effort: "max". Smaller contexts fall back to high reasoning.

The first request is slow

ds4 has a cold Metal residency and model warmup phase. Use localService.readyTimeoutMs: 300000 when OpenClaw starts the server on demand.

如果把这一段摆成一个小场景，你会看到几样东西正在互相打招呼、拦路或者传东西。别急着记名词，先抓住它此刻到底在发生什么：Card大家一起的房间 Local 聪明脑袋模型 services Start local 聪明脑袋模型 大房子服务器s on demand before 聪明脑袋模型 requests. Local 聪明脑袋模型s Choose and…。

CardGroup

Local model services

Start local model servers on demand before model requests.

Local models

Choose and operate local model backends.

Model providers

Configure provider refs, auth, and failover.

DeepSeek

Native DeepSeek provider behavior and thinking controls.

---