Mintplex-Labs/anything-llm · error

No Ollama image generation base path was set.

Error message

No Ollama image generation base path was set.

What it means

The OllamaImageGenerator constructor throws immediately when IMAGE_GEN_OLLAMA_BASE_PATH is unset or empty. The value is normalized (trailing slashes stripped) and suffixed with /v1 to build the OpenAI-compatible client base URL — Ollama serves experimental image generation through that endpoint. Fails fast at instantiation.

Solutions

1. Set IMAGE_GEN_OLLAMA_BASE_PATH to the Ollama server root, e.g. http://localhost:11434 (the code appends /v1 itself).
2. From docker, use http://host.docker.internal:11434 instead of localhost.
3. Restart the server / recreate the container after saving.
4. Confirm your Ollama build actually exposes the OpenAI-compatible /v1/images endpoint (experimental, macOS).

Example fix

# .env — before
IMAGE_GEN_ENGINE=ollama

# .env — after
IMAGE_GEN_ENGINE=ollama
IMAGE_GEN_OLLAMA_BASE_PATH=http://localhost:11434

if (!process.env.IMAGE_GEN_OLLAMA_BASE_PATH) {
  throw new Error(
    "Ollama image engine selected but IMAGE_GEN_OLLAMA_BASE_PATH is missing."
  );
}
const { OllamaImageGenerator } = require("server/utils/ImageGenerators/ollama");
const generator = new OllamaImageGenerator();

function isOllamaImageConfigValid(env = process.env) {
  return (
    typeof env.IMAGE_GEN_OLLAMA_BASE_PATH === "string" &&
    env.IMAGE_GEN_OLLAMA_BASE_PATH.length > 0
  );
}

try {
  new OllamaImageGenerator();
} catch (err) {
  if (err.message.includes("Ollama image generation base path")) {
    // surface a settings error for IMAGE_GEN_OLLAMA_BASE_PATH
  }
}

When it happens

Trigger: Selecting Ollama as the image engine without setting IMAGE_GEN_OLLAMA_BASE_PATH; docker deployments where localhost would point at the container rather than the host; .env updated without a restart; typo in the variable name.

Common situations: Trying Ollama's experimental image generation (macOS-only per the code comment) for the first time; reusing the chat OLLAMA basePath variable but not the image-specific one.

Understand the failure class

Background: "environment variable is not set" and "Missing keys in environment" errors: what missing required env var messages mean and how to fix them — this error's family across 28 libraries.

Related errors

• No Ollama image generation model was set.
• No embedding model was set.
• No Lemonade image generation base path was set.
• No Lemonade image generation model was set.
• No OpenAI image generation API key was set.

AI-assisted analysis of Mintplex-Labs/anything-llm@20f6d3546c (2026-08-18). Data as JSON: /api/errors/84d165890e360444. Report an issue: GitHub.

Appendix: source

Thrown at server/utils/ImageGenerators/ollama/index.js:6

const { BaseImageGenerator } = require("../base");

class OllamaImageGenerator extends BaseImageGenerator {
  constructor() {
    if (!process.env.IMAGE_GEN_OLLAMA_BASE_PATH)
      throw new Error("No Ollama image generation base path was set.");
    if (!process.env.IMAGE_GEN_MODEL_PREF)
      throw new Error("No Ollama image generation model was set.");
    const { OpenAI: OpenAIApi } = require("openai");
    // Ollama serves image generation through its OpenAI-compatible `/v1`
    // endpoint (experimental, macOS only).
    const basePath = process.env.IMAGE_GEN_OLLAMA_BASE_PATH.replace(/\/+$/, "");
    super({
      client: new OpenAIApi({
        baseURL: `${basePath}/v1`,
        apiKey: process.env.IMAGE_GEN_OLLAMA_AUTH_TOKEN || "ollama",
      }),
      model: process.env.IMAGE_GEN_MODEL_PREF,
      className: "OllamaImageGenerator",
    });
  }

  /**
   * Ollama's OpenAI-compatible image endpoint does not implement image editing

View on GitHub (pinned to 20f6d3546c)