continuedev/continue · error · Error

'${modelName}' not found in the Ollama registry!

Error message

'${modelName}' not found in the Ollama registry!

What it means

Thrown by Ollama.installModel when getRemoteModelInfo returns null, i.e. the model name couldn't be found in the Ollama registry lookup. The name must match a registry tag exactly (including namespaces and tag suffixes).

Source

Thrown at core/llm/llms/Ollama.ts:767

    }

    const data = await resp.json();
    const embedding: number[][] = data.embeddings;

    if (!embedding || embedding.length === 0) {
      throw new Error("Ollama generated empty embedding");
    }
    return embedding;
  }

  public async installModel(
    modelName: string,
    signal: AbortSignal,
    progressReporter?: (task: string, increment: number, total: number) => void,
  ): Promise<any> {
    const modelInfo = await getRemoteModelInfo(modelName, signal);
    if (!modelInfo) {
      throw new Error(`'${modelName}' not found in the Ollama registry!`);
    }

    const release = await Ollama.modelsBeingInstalledMutex.acquire();
    try {
      if (Ollama.modelsBeingInstalled.has(modelName)) {
        throw new Error(`Model '${modelName}' is already being installed.`);
      }
      Ollama.modelsBeingInstalled.add(modelName);
    } finally {
      release();
    }

    try {
      const response = await fetch(this.getEndpoint("api/pull"), {
        method: "POST",
        headers: {
          "Content-Type": "application/json",
          Authorization: `Bearer ${this.apiKey}`,

View on GitHub (pinned to 5522c6f44c)

Solutions

1. Search https://ollama.com/library for the exact name+tag and retry
2. Use the plain library name (e.g. 'qwen2.5-coder:7b'), not a HF repo id
3. If the name is right, check network access to the Ollama registry

Example fix

# before
installModel('meta-llama/Llama-3.1-8B-Instruct')
# after
installModel('llama3.1:8b')

const info = await getRemoteModelInfo(modelName, signal);
if (!info) throw new Error(`Check spelling at https://ollama.com/library: ${modelName}`);

function isRegistryLookupFailure(e: unknown): boolean { return e instanceof Error && /not found in the Ollama registry/.test(e.message); }

try { await Ollama.installModel(name, signal); }
catch (e) { if (isRegistryLookupFailure(e)) promptUserForExactName(); else throw e; }

When it happens

Trigger: Calling installModel with a name that has no registry entry: typo, missing tag (e.g. 'llama3.1' vs 'llama3.1:8b'), or a registry API/network failure that yields no info.

Common situations: User typos in the model name, using HuggingFace-style ids (meta-llama/Llama-3.1-8B) instead of Ollama library names, or ollama.com being unreachable.

Related errors

• Model '${modelName}' is already being installed.
• Block ${packageIdentifierToShorthandSlug(id)} not found
• FindAndReplaceOldStringNotFound
• Failed to fetch Ollama library: ${response.status}
• Error fetching tags: ${e.message}

AI-assisted analysis of continuedev/continue@5522c6f44c (2026-08-27). Data as JSON: /api/errors/dd770486c790e6fc. Report an issue: GitHub.