Mintplex-Labs/anything-llm · error · Error

${e.message}

Error message

${e.message}

What it means

Re-thrown from the OpenAI SDK rejection inside SambaNovaLLM.getChatCompletion via `.catch((e) => { throw new Error(e.message); })`. SambaNova does NOT pre-validate the model (isValidChatCompletionModel just returns !!modelName), so unlike Perplexity/PPIO/TogetherAI, an unknown-model error surfaces here as a runtime API failure rather than a pre-flight throw. The wrapper strips the SDK error class/status, leaving only the message.

Source

Thrown at server/utils/AiProviders/sambanova/index.js:125

      prompt,
      ...chatHistory,
      {
        role: "user",
        content: this.#generateContent({ userPrompt, attachments }),
      },
    ];
  }

  async getChatCompletion(messages = null, { temperature = 0.7 }) {
    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
    )
      return null;

    return {
      textResponse: result.output.choices[0].message.content,
      metrics: {
        prompt_tokens: result.output.usage?.prompt_tokens || 0,
        completion_tokens: result.output.usage?.completion_tokens || 0,
        total_tokens: result.output.usage?.total_tokens || 0,
        outputTps: result.output.usage?.total_tokens_per_sec || 0,
        duration: result.duration,
        model: this.model,

View on GitHub (pinned to 526360e320)

Solutions

1. If the message mentions the model: switch SAMBANOVA_LLM_MODEL_PREF to a model the key is authorised to call (check the SambaNova console's enabled-models list).
2. curl POST https://api.sambanova.ai/v1/chat/completions with the same key/model to isolate auth vs model vs quota.
3. On 429, throttle or add bounded exponential-backoff retry.
4. On 401, rotate the key and restart.

Example fix

// before
const out = await llm.getChatCompletion(messages, { temperature: 0.7 });

// after
const out = await (async () => {
  for (let i = 0; i < 3; i++) {
    try {
      return await llm.getChatCompletion(messages, { temperature: 0.7 });
    } catch (e) {
      if (/401|not authorized|model/i.test(e.message)) throw e; // non-transient
      if (i === 2) throw e;
      await new Promise((r) => setTimeout(r, 2 ** i * 500));
    }
  }
})();

function classifySambanovaError(message) {
  if (/401|unauthorized|invalid api key/i.test(message)) return "auth";
  if (/model|not found|not authorized|deployed/i.test(message)) return "model";
  if (/429|rate limit|quota/i.test(message)) return "rate";
  if (/5\d{2}|timeout|ECONN/i.test(message)) return "transient";
  return "fatal";
}
// SambaNova does no client-side model check, so model/auth errors arrive here

function isTransientSambanovaError(message) {
  return typeof message === "string" && /429|5\d{2}|timeout|ECONNRESET|fetch failed/i.test(message);
}

async function sambanovaCall(llm, messages, opts, retries = 3) {
  for (let i = 0; i <= retries; i++) {
    try {
      return await llm.getChatCompletion(messages, opts);
    } catch (e) {
      const kind = classifySambanovaError(e.message);
      if (kind === "auth" || kind === "model" || i === retries) throw e;
      await new Promise((r) => setTimeout(r, 2 ** i * 500));
    }
  }
}

When it happens

Trigger: api.sambanova.ai returns non-2xx for the create call: 401 (bad key), 400 (model id not served by this key / not deployed), 429 (rate limit), 5xx, or a transport/abort error before a response.

Common situations: SAMBANOVA_LLM_MODEL_PREF references a model not enabled for the API key's tenant (SambaNova gates models per deployment); key revoked after startup; burst traffic hit the rate ceiling; cold start timeout; client aborted mid-request.

Related errors

• ${e.message}
• ${e.message}
• ${e.message}
• ${e.message}
• e.message

AI-assisted analysis of Mintplex-Labs/anything-llm@526360e320 (2026-08-13). Data as JSON: /api/errors/e402fec001d771f8. Report an issue: GitHub.