eyaltoledano/claude-task-master · error

Base URL is required for OpenAI-compatible providers. Please

Error message

Base URL is required for OpenAI-compatible providers. Please provide a baseURL.

What it means

Thrown by setModel when the user sets a model with the --openai-compatible provider hint but no baseURL can be resolved. Unlike LM Studio, OpenAI-compatible providers have no sensible default endpoint, so the library refuses to save a config entry that would be unusable. The base URL must come from the --base-url flag or from an existing baseURL already stored for that role while the provider was already OPENAI_COMPATIBLE.

Source

Thrown at scripts/modules/task-manager/models.js:668

					// Get current provider for this role to check if we should preserve baseURL
					let currentProvider;
					if (role === 'main') {
						currentProvider = getMainProvider(projectRoot);
					} else if (role === 'research') {
						currentProvider = getResearchProvider(projectRoot);
					} else if (role === 'fallback') {
						currentProvider = getFallbackProvider(projectRoot);
					}

					// Only preserve baseURL if we're already using OPENAI_COMPATIBLE
					const existingBaseURL =
						currentProvider === CUSTOM_PROVIDERS.OPENAI_COMPATIBLE
							? getBaseUrlForRole(role, projectRoot)
							: null;

					const resolvedBaseURL = baseURL || existingBaseURL;
					if (!resolvedBaseURL) {
						throw new Error(
							`Base URL is required for OpenAI-compatible providers. Please provide a baseURL.`
						);
					}
					warningMessage = `Warning: Custom OpenAI-compatible model '${modelId}' set with base URL '${resolvedBaseURL}'. Taskmaster cannot guarantee compatibility. Ensure your API endpoint follows the OpenAI API specification.`;
					report('warn', warningMessage);
					// Store the computed baseURL so it gets saved in config
					computedBaseURL = resolvedBaseURL;
				} else {
					// Invalid provider hint - should not happen with our constants
					throw new Error(`Invalid provider hint received: ${providerHint}`);
				}
			}
		} else {
			// No hint provided (flags not used)
			if (modelData) {
				// Found internally, use the provider from the internal list
				determinedProvider = modelData.provider;
				report(

View on GitHub (pinned to c0c98d367c)

Solutions

1. Pass the endpoint with --base-url, e.g. `task-master models --set-main=my-model --openai-compatible --base-url=https://my-endpoint/v1`
2. If the role already used an OpenAI-compatible provider, verify its baseURL exists in .taskmaster/config.json under models.<role>.baseURL so it can be preserved; otherwise set it explicitly
3. Use the LM Studio hint instead (it defaults to http://localhost:1234/v1) only if that is actually your setup
4. Programmatically, call setModel with baseURL in the options object

Example fix

// before
task-master models --set-main=llama-3-70b --openai-compatible
// after
task-master models --set-main=llama-3-70b --openai-compatible --base-url=https://api.together.xyz/v1

import { getBaseUrlForRole } from './utils.js';
function assertBaseURL(baseURL, role, projectRoot, currentProvider, CUSTOM_PROVIDERS) {
  const existing = currentProvider === CUSTOM_PROVIDERS.OPENAI_COMPATIBLE
    ? getBaseUrlForRole(role, projectRoot) : null;
  if (!baseURL && !existing) {
    throw new Error(`--base-url is required when setting an OpenAI-compatible model for role '${role}'`);
  }
}

function hasBaseUrl(opts) {
  return typeof opts.baseURL === 'string' && opts.baseURL.trim().length > 0;
}

try {
  await setModel(projectRoot, role, modelId, { providerHint: 'OPENAI_COMPATIBLE', baseURL });
} catch (err) {
  if (/Base URL is required/i.test(err.message)) {
    console.error('Provide --base-url, e.g. --base-url=https://your-endpoint/v1');
  } else throw err;
}

When it happens

Trigger: Calling setModel (or `task-master models --set-main=<id> --openai-compatible`) with a custom model ID and the OPENAI_COMPATIBLE provider hint, where the --base-url flag is omitted AND the role's current provider is not already OPENAI_COMPATIBLE (so no existingBaseURL is preserved).

Common situations: First-time setup of a custom OpenAI-compatible endpoint (vLLM, LiteLLM proxy, Together, etc.) where the user forgot --base-url; switching from another provider (e.g. openrouter) to openai-compatible and assuming the old baseURL carries over — it deliberately does not; scripting the models command and dropping the flag.

Related errors

• Invalid provider hint received: ${providerHint}
• ${this.name} API key is required
• ${this.name} Model ID is required
• Provider name is required
• API key environment variable name is required

AI-assisted analysis of eyaltoledano/claude-task-master@c0c98d367c (2026-08-29). Data as JSON: /api/errors/b966d69f35989585. Report an issue: GitHub.