mem0ai/mem0 · error · Error

Sarvam LLM failed: ${message}

Error message

Sarvam LLM failed: ${message}

What it means

Thrown by SarvamLLM.generateResponse when the underlying request to Sarvam's OpenAI-compatible endpoint (default https://api.sarvam.ai/v1) fails. The suffix after 'Sarvam LLM failed:' preserves the raw Sarvam API error — typically 401 invalid key, 429 rate limit, subscription/quota issues, or invalid model.

Source

Thrown at mem0-ts/src/oss/src/llms/sarvam.ts:40

      apiKey,
      baseURL:
        config.baseURL ||
        process.env.SARVAM_API_BASE ||
        "https://api.sarvam.ai/v1",
      model: config.model || "sarvam-m",
    });
  }

  async generateResponse(
    messages: Message[],
    responseFormat?: { type: string },
    tools?: any[],
  ): Promise<string | LLMResponse> {
    try {
      return await super.generateResponse(messages, responseFormat, tools);
    } catch (err) {
      const message = err instanceof Error ? err.message : String(err);
      throw new Error(`Sarvam LLM failed: ${message}`);
    }
  }

  async generateChat(messages: Message[]): Promise<LLMResponse> {
    try {
      return await super.generateChat(messages);
    } catch (err) {
      const message = err instanceof Error ? err.message : String(err);
      throw new Error(`Sarvam LLM failed: ${message}`);
    }
  }
}

View on GitHub (pinned to 001c235229)

Solutions

1. Read the suffix: 401 → fix SARVAM_API_KEY; 429/quota → back off or top up the Sarvam plan; invalid model → correct the identifier.
2. Verify the key with a direct call: curl https://api.sarvam.ai/v1/models -H "Authorization: Bearer $SARVAM_API_KEY".
3. Retry transient (429/5xx/network) failures with exponential backoff; treat 401 as fatal.
4. If routing through a custom base, confirm the gateway is up and OpenAI-compatible.

Example fix

// before
await sarvamLlm.generateResponse(messages, { type: 'json_object' });
// Sarvam LLM failed: 401 Unauthorized

// after
try {
  await sarvamLlm.generateResponse(messages, { type: 'json_object' });
} catch (e) {
  if (/401/.test(String(e))) throw new ConfigError('Sarvam key invalid');
  throw e;
}

async function sarvamAuthed(key = process.env.SARVAM_API_KEY, base = 'https://api.sarvam.ai/v1') {
  const r = await fetch(`${base}/models`, { headers: { Authorization: `Bearer ${key}` } });
  if (r.status === 401) throw new Error('Sarvam key invalid');
  return r.ok;
}

const isSarvamFatal = (e: unknown): boolean =>
  e instanceof Error && e.message.startsWith('Sarvam LLM failed:') && /401|quota|credit/i.test(e.message);

try {
  return await sarvamLlm.generateResponse(messages, responseFormat);
} catch (err) {
  if (isSarvamFatal(err)) throw err;
  if (/429|timeout|ECONN/i.test(String(err))) return withRetry(fn, { retries: 3 });
  throw err;
}

When it happens

Trigger: Calling generateResponse() with an invalid/expired Sarvam key, exhausted API credits, a wrong model name (default 'sarvam-m'), an unsupported responseFormat for the model, or when api.sarvam.ai / a custom SARVAM_API_BASE is unreachable.

Common situations: See trigger scenarios.

Related errors

• MiniMax LLM failed: ${message}
• Sarvam API key is required
• DeepSeek API key is required
• DeepSeek LLM failed: ${message}
• Groq API key is required

AI-assisted analysis of mem0ai/mem0@001c235229 (2026-08-15). Data as JSON: /api/errors/0f679ed7ccee4b03. Report an issue: GitHub.