ruvnet/ruflo · error · LLMProviderError

OLLAMA_${response.status}

OLLAMA_${response.status}

Error message

${message}

What it means

OllamaProvider's fallback for non-connection errors: LLMProviderError with code OLLAMA_<status>, and unlike the cloud providers it hardcodes retryable=true for every status. The message is Ollama's JSON error field ('Unknown error' if not JSON).

Source

Thrown at v3/@claude-flow/providers/src/ollama-provider.ts:399

    const errorText = await response.text();
    let errorData: { error?: string };

    try {
      errorData = JSON.parse(errorText);
    } catch {
      errorData = { error: errorText };
    }

    const message = errorData.error || 'Unknown error';

    if (response.status === 0 || message.includes('connection')) {
      throw new ProviderUnavailableError('ollama', {
        message,
        hint: 'Ensure Ollama is running: ollama serve',
      });
    }

    throw new LLMProviderError(
      message,
      `OLLAMA_${response.status}`,
      'ollama',
      response.status,
      true,
      errorData
    );
  }
}

View on GitHub (pinned to fa13ee4ad6)

Solutions

1. Run `ollama list` and use an exact tag from it, pulling if needed: `ollama pull llama3.1:8b`
2. Read error.details.error - Ollama's message names the concrete cause (model missing, bad parameter, server error)
3. For 5xx: retry with backoff; if it persists restart the daemon or upgrade Ollama to a version that supports the model

Example fix

// before
config: { model: 'llama3.1', apiUrl: 'http://localhost:11434' } // tag not pulled -> OLLAMA_404

// after
// $ ollama pull llama3.1:8b
config: { model: 'llama3.1:8b', apiUrl: 'http://localhost:11434' } // matches `ollama list` output

async function ollamaHasModel(model: string, apiUrl = 'http://localhost:11434'): Promise<boolean> {
  const r = await fetch(`${apiUrl}/api/tags`);
  const { models } = await r.json();
  return (models ?? []).some((m: { name: string }) => m.name === model);
}

import { LLMProviderError, isLLMProviderError } from './types.js';
function isOllamaApiError(e: unknown): e is LLMProviderError {
  return isLLMProviderError(e) && e.provider === 'ollama' && e.code.startsWith('OLLAMA_');
}

try {
  return await provider.complete(req);
} catch (e) {
  if (isOllamaApiError(e)) {
    // note: this mapping marks every status retryable; still inspect it first
    if (e.statusCode === 404) throw new Error('model not pulled locally: run ollama pull ' + req.model);
    return retryWithBackoff(() => provider.complete(req));
  }
  throw e;
}

When it happens

Trigger: complete() when the requested model tag has not been pulled (404 "model 'x' not found"), malformed request bodies (400), or Ollama server errors (500) - any non-connection HTTP response.

Common situations: Model tag typo or missing tag ('llama3.1' vs 'llama3.1:8b'); model pulled under a different name on another machine; Ollama version too old for a newer model/feature; corrupted model store.

Understand the failure class

Background: "API error: {status}" and "HTTP 401/403/404/429/5xx" errors: non-2xx HTTP responses explained — this error's family across 27 libraries.

Related errors

• PROVIDER_UNAVAILABLE
• OLLAMA_${response.status}
• MCP server "${server.name}" is in cooldown (HTTP ${cd.status
• COHERE_${response.status}
• GOOGLE_${response.status}

AI-assisted analysis of ruvnet/ruflo@fa13ee4ad6 (2026-08-18). Data as JSON: /api/errors/3902de8a7e4878f9. Report an issue: GitHub.