Mintplex-Labs/anything-llm · error · Error

No Privatemode Base Path was set.

Error message

No Privatemode Base Path was set.

What it means

Thrown by the PrivatemodeLLM constructor when process.env.PRIVATEMODE_LLM_BASE_PATH is falsy. Privatemode is a self-hosted/local OpenAI-compatible server, so its base URL is mandatory; there is no default endpoint. The constructor then runs the value through parseBasePath() which appends /v1.

Source

Thrown at server/utils/AiProviders/privatemode/index.js:21

  handleDefaultStreamResponseV2,
  formatChatHistory,
} = require("../../helpers/chat/responses");
const {
  LLMPerformanceMonitor,
} = require("../../helpers/chat/LLMPerformanceMonitor");

class PrivatemodeLLM {
  static contextWindows = {
    "leon-se/gemma-3-27b-it-fp8-dynamic": 128000,
    "gemma-3-27b": 128000,
    "qwen3-coder-30b-a3b": 128000,
    "gpt-oss-120b": 128000,
    "openai/gpt-oss-120b": 128000,
  };

  constructor(embedder = null, modelPreference = null) {
    if (!process.env.PRIVATEMODE_LLM_BASE_PATH)
      throw new Error("No Privatemode Base Path was set.");

    this.className = "PrivatemodeLLM";
    const { OpenAI: OpenAIApi } = require("openai");
    this.client = new OpenAIApi({
      baseURL: PrivatemodeLLM.parseBasePath(),
      apiKey: null,
    });

    this.model = modelPreference || process.env.PRIVATEMODE_LLM_MODEL_PREF;
    this.limits = {
      history: this.promptWindowLimit() * 0.15,
      system: this.promptWindowLimit() * 0.15,
      user: this.promptWindowLimit() * 0.7,
    };

    this.embedder = embedder ?? new NativeEmbedder();
    this.defaultTemp = 0.7;
    this.log(

View on GitHub (pinned to 526360e320)

Solutions

1. Set PRIVATEMODE_LLM_BASE_PATH to the Privatemode server origin (e.g. http://127.0.0.1:8000) in server/.env and restart.
2. Confirm the Privatemode process is listening: `curl -s http://127.0.0.1:8000/v1/models`.
3. If behind Docker networking, use the service name/host reachable from the AnythingLLM container.
4. Restart AnythingLLM so the new env is loaded.

Example fix

// before
// PRIVATEMODE_LLM_BASE_PATH unset -> throws
const llm = new PrivatemodeLLM(embedder, "gemma-3-27b");

// after
// server/.env
// PRIVATEMODE_LLM_BASE_PATH=http://127.0.0.1:8000
// PRIVATEMODE_LLM_MODEL_PREF=gemma-3-27b
const llm = new PrivatemodeLLM(embedder, "gemma-3-27b");

function assertPrivatemodeBasePath() {
  if (!process.env.PRIVATEMODE_LLM_BASE_PATH || !process.env.PRIVATEMODE_LLM_BASE_PATH.trim()) {
    throw new Error("PRIVATEMODE_LLM_BASE_PATH missing — set to your Privatemode server origin (e.g. http://127.0.0.1:8000)");
  }
}
assertPrivatemodeBasePath();
const llm = new PrivatemodeLLM(embedder, modelPref);

function hasPrivatemodeBasePath(env = process.env) {
  return typeof env.PRIVATEMODE_LLM_BASE_PATH === "string" && env.PRIVATEMODE_LLM_BASE_PATH.trim().length > 0;
}

let llm;
try {
  llm = new PrivatemodeLLM(embedder, modelPref);
} catch (e) {
  if (/No Privatemode Base Path/i.test(e.message)) {
    return { ok: false, reason: "missing-base-path" };
  }
  throw e;
}

When it happens

Trigger: Instantiating `new PrivatemodeLLM(...)` with PRIVATEMODE_LLM_BASE_PATH unset, empty, or whitespace-only. The variable must be the origin (and optionally path) of the local Privatemode inference server, e.g. http://127.0.0.1:8000.

Common situations: Privatemode container/process not yet started when AnythingLLM boots; the variable was named PRIVATEMODE_BASE_PATH instead of PRIVATEMODE_LLM_BASE_PATH; the env file was not reloaded after adding it; a docker-compose override omitted the mapping.

Related errors

• TextGenWebUI must have a valid base path to use for the api.
• No NVIDIA NIM API Base Path was set.
• No Perplexity API key was set.
• No PPIO API key was set.
• No SambaNova API key was set.

AI-assisted analysis of Mintplex-Labs/anything-llm@526360e320 (2026-08-13). Data as JSON: /api/errors/22b4eceb5654accc. Report an issue: GitHub.