Mintplex-Labs/anything-llm · error · Error

Novita chat: is not valid for chat completion!

Error message

Novita chat: ${this.model} is not valid for chat completion!

What it means

NovitaLLM.getChatCompletion validates this.model by first syncing the Novita model catalog (#syncModels, cached on disk under storage/models/novita) and then checking whether the model id is a key of this.models(). If the configured id is not in the synced catalog, the check fails and this error is thrown before any completion request.

Solutions

  1. Force a model re-sync: delete storage/models/novita (respecting STORAGE_DIR if set) or refresh the model list in the AnythingLLM UI, then retry
  2. Verify the exact id via the Novita API model list and set NOVITA_LLM_MODEL_PREF to it
  3. Confirm network egress to api.novita.ai works — the 3s sync timeout silently produces an empty catalog
  4. Re-select the model from the UI dropdown after the list refreshes

Example fix

# before
NOVITA_LLM_MODEL_PREF=meta-llama/old-id
# after (id must appear in the synced Novita catalog)
NOVITA_LLM_MODEL_PREF=meta-llama/llama-3.1-8b-instruct
Defensive patterns

Strategy: validation

Validate before calling

const models = llm.models(); // populated by #syncModels()
if (!Object.prototype.hasOwnProperty.call(models, llm.model))
  throw new Error(`Model ${llm.model} not in synced Novita catalog — refresh model list`);

Try / catch

try {
  await llm.getChatCompletion(messages);
} catch (e) {
  if (/Novita chat: .* is not valid/.test(e.message)) {
    // refresh storage/models/novita cache, confirm the id, then retry
  }
  throw e;
}

Prevention

When it happens

Trigger: NOVITA_LLM_MODEL_PREF or a workspace-selected model id that Novita no longer offers (renamed/deprecated); a stale or empty local model cache because the sync request failed (NovitaLLM sets a 3s default timeout, so slow networks can yield an empty catalog); model ids copied from outdated docs.

Common situations: Novita rotating its catalog while the on-disk cache is stale; picking model ids from old tutorials; flaky egress making the initial sync return nothing.

Related errors

AI-assisted analysis of Mintplex-Labs/anything-llm@3aec848f28 (2026-08-18). Data as JSON: /api/errors/3f8380d7aab958d6. Report an issue: GitHub.

Appendix: source

Thrown at server/utils/AiProviders/novita/index.js:224

    attachments = [],
  }) {
    const prompt = {
      role: "system",
      content: `${systemPrompt}${this.#appendContext(contextTexts)}`,
    };
    return [
      prompt,
      ...formatChatHistory(chatHistory, this.#generateContent),
      {
        role: "user",
        content: this.#generateContent({ userPrompt, attachments }),
      },
    ];
  }

  async getChatCompletion(messages = null, { temperature = 0.7 }) {
    if (!(await this.isValidChatCompletionModel(this.model)))
      throw new Error(
        `Novita chat: ${this.model} is not valid for chat completion!`
      );

    const result = await LLMPerformanceMonitor.measureAsyncFunction(
      this.openai.chat.completions
        .create({
          model: this.model,
          messages,
          temperature,
        })
        .catch((e) => {
          throw new Error(e.message);
        })
    );

    if (
      !result.output.hasOwnProperty("choices") ||
      result.output.choices.length === 0

View on GitHub (pinned to 3aec848f28)