Mintplex-Labs/anything-llm · error · RetryError

error.message

Error message

error.message

What it means

Cohere provider's `stream()` classifies OpenAI SDK errors. It logs `console.error(error.message, error)` first (so the full error object is in the server logs), then re-throws `AuthenticationError`, wraps `RateLimitError`/`InternalServerError`/`APIError` as `RetryError(error.message)`, and re-throws everything else. Cohere is reached through an OpenAI-compatible endpoint.

Source

Thrown at server/utils/agents/aibitat/providers/cohere.js:116

    try {
      return await tooledStream(
        this.client,
        this.model,
        messages,
        functions,
        eventHandler,
        { provider: this }
      );
    } catch (error) {
      console.error(error.message, error);
      if (error instanceof OpenAI.AuthenticationError) throw error;
      if (
        error instanceof OpenAI.RateLimitError ||
        error instanceof OpenAI.InternalServerError ||
        error instanceof OpenAI.APIError
      ) {
        throw new RetryError(error.message);
      }
      throw error;
    }
  }

  async complete(messages, functions = []) {
    const useNative = await this.supportsNativeToolCalling();

    if (!useNative) {
      return await UnTooled.prototype.complete.call(
        this,
        messages,
        functions,
        this.#handleFunctionCallChat.bind(this)
      );
    }

    try {

View on GitHub (pinned to 526360e320)

Solutions

1. Inspect the server console — stream errors are console.error'd with the full object before rethrow.
2. Treat RetryError as transient: let the framework retry the turn.
3. On repeated 429, verify the COHERE_API_KEY tier and request volume; upgrade or throttle.
4. Confirm the model slug is a current Cohere model (command-r, command-r-plus, aya, etc.).
5. Check the Cohere status page for incidents.

// Verify a Cohere model slug is current before streaming.
const KNOWN_COHERE_MODELS = new Set(['command-r','command-r-plus','aya','command-r-08-2024','command-r-plus-08-2024']);
function assertCohereModel(model) {
  if (!KNOWN_COHERE_MODELS.has(model)) throw new Error(`Unknown Cohere model: ${model}`);
}

function isRetryableProviderError(e) {
  return e instanceof OpenAI.RateLimitError || e instanceof OpenAI.InternalServerError || e instanceof OpenAI.APIError;
}

// Stream path logs to console.error — read it. Retry on RetryError, fail on auth.
try { return await provider.stream(messages, functions, handler); }
catch (e) {
  if (e instanceof OpenAI.AuthenticationError) throw e;
  if (e instanceof RetryError) return await backoffAndRetry();
  throw e;
}

When it happens

Trigger: Cohere 429 (trial/production key rate ceiling); Cohere 5xx; APIError from a model Cohere rejected (e.g. a deprecated command/aya slug); stream interrupted mid-flight and surfaced by the SDK as APIError.

Common situations: Cohere trial key quota exhausted; a non-Cohere model slug sent to the Cohere endpoint; regional Cohere outage; streaming connection dropped by an intermediary proxy.

Related errors

• error.message
• error.message
• error.message
• error.message
• error.message

AI-assisted analysis of Mintplex-Labs/anything-llm@526360e320 (2026-08-13). Data as JSON: /api/errors/e5bd5b93a56f548f. Report an issue: GitHub.