Neura Market Logo

ds4 Provider: Run DeepSeek V4 Flash Locally with OpenClaw

This page explains how to configure OpenClaw to use ds4, a local Metal backend that serves DeepSeek V4 Flash via an OpenAI-compatible API. It covers setup, provider configuration, and model selection for users running ds4 on macOS.

Read this when

  • You want to run OpenClaw against antirez/ds4
  • You want a local DeepSeek V4 Flash backend with tool calls
  • You need the OpenClaw config for ds4-server

ds4 runs DeepSeek V4 Flash from a local Metal backend through an OpenAI-compatible /v1 API. OpenClaw connects to ds4 using 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.

PropertyValue
Provider idds4
Pluginnone (config-only)
APIOpenAI-compatible Chat Completions (openai-completions)
Base URLhttp://127.0.0.1:18000/v1 (suggested)
Model iddeepseek-v4-flash
Tool callsOpenAI-style tools / tool_calls
ReasoningDeepSeek-style thinking and reasoning_effort

Requirements

  • 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 KV memory at server startup.

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 with enough memory and to enable ds4 Think Max.

Quickstart

Start ds4-server

Replace <DS4_DIR> with your ds4 checkout path.

<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

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:

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

Full config

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

On-demand startup

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.

Think Max

ds4 applies Think Max only when both 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"],
  },
}

Test

Direct HTTP check, bypassing OpenClaw:

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"}}'

OpenClaw model routing (same as the Quickstart check):

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

Full agent and tool-call smoke test, with 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

Troubleshooting

curl /v1/models cannot connect

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

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. Set localService.readyTimeoutMs: 300000 when OpenClaw starts the server on demand.