Sarvam AI · Rate Limits

Sarvam Ai Rate Limits

Sarvam AI enforces per-account rate limits expressed primarily as requests per minute (RPM), applied per product (chat LLM, speech-to-text, text-to-speech, translate). Limits are enforced at the account level - all API keys on an account share the same pool - and rise with plan tier (Starter, Pro, Business, Enterprise). The Starter plan allows 60 RPM for the chat LLM API. Specific per-product values for higher tiers are not reconciled in this artifact.

Sarvam Ai Rate Limits is the machine-readable rate-limit profile for Sarvam AI on the APIs.io network, conforming to the API Commons Rate Limits specification.

It captures 4 rate-limit definitions, measuring requests.

The profile also includes 3 backoff/retry policies defined and response codes documented for throttled.

Tagged areas include AI, LLM, Speech to Text, Text to Speech, and Translation.

4 Limits Throttle: 429
AILLMSpeech to TextText to SpeechTranslationIndian LanguagesRate LimitingQuotasThrottling

Limits

Requests Per Minute (RPM) - Chat LLM account
requests
60 on Starter; higher on paid tiers
Per-account RPM for the chat completions API; shared across all keys.
Requests Per Minute (RPM) - Speech-to-Text account
requests
see provider dashboard
Per-account RPM for speech-to-text and speech-to-text-translate; varies by tier.
Requests Per Minute (RPM) - Text-to-Speech account
requests
see provider dashboard
Per-account RPM for text-to-speech; varies by tier.
Requests Per Minute (RPM) - Translate / Transliterate account
requests
see provider dashboard
Per-account RPM for translate, transliterate, and language ID; varies by tier.

Policies

Account-Level Enforcement
Limits apply to the account as a whole; all API keys share the same rate-limit pool.
Tiered Limits
Limits raise as accounts move from Starter to Pro, Business, and Enterprise plans.
Backoff Strategy
Clients should implement exponential backoff with jitter and honor Retry-After on 429 responses.

Sources