---
title: "Sarvam AI free inference"
description: "₹100 for new accounts; it does not recur. See verified limits, model IDs, privacy terms, and primary sources for Sarvam AI."
canonical_url: "https://freeinferencing.com/provider/sarvam_ai/"
md_url: "https://freeinferencing.com/provider/sarvam_ai.md"
last_updated: "2026-09-19"
---

# Sarvam AI

> ₹100 for new accounts; it does not recur.

## Classification

- **Directory:** Current
- **Free access:** One-time trial
- **Status:** Current trial
- **Confidence:** High
- **Payment card:** No
- **Account:** Required
- **Equivalent paid value:** ₹100.00 once
- **API endpoint:** `https://api.sarvam.ai/v2`

## Models mentioned

- `glm5.3`
- `gemma4`
- `sarvam-105b`
- `deepseekv4-flash`
- `deepseekv4.1-flash`

## Limits and terms

```yaml
signup_credit_inr: 100
expiry: none
recurrence: false
starter_requests_per_minute: 60
```

## What happens to your prompts?

**Not private.** Free developer inputs and outputs are collected and may support service improvement without a clear default no-training commitment.

Privacy is audited separately from price. Review the current governing terms before sending sensitive or regulated data.

## Data governance and agreements

```yaml
data_governance:
  review_status: partial
  plan_scope: Sarvam developer APIs; enterprise customer content can be governed
    by separate customer agreements.
  prompt_retention: The public privacy policy says the products collect user
    inputs, file uploads, and generated outputs, but does not publish a
    developer-API content TTL.
  response_retention: Outputs are included in collected product information
    without a fixed public retention period.
  ordinary_logging: Standard usage logs, IP/device data, traffic, cookies,
    diagnostics, and product use are collected.
  model_training: No clear default developer-API commitment excluding inputs and
    outputs from model training was found; enterprise agreements may differ.
  product_improvement: The policy permits data use to enhance user experience and
    improve services; beta/trial terms expressly allow diagnostic, performance,
    and usage analysis.
  human_or_operator_access: Sarvam and service providers may process collected
    content for delivery, support, safety, and legal purposes.
  subprocessors_and_routing: Service providers and infrastructure vendors may
    receive necessary data; enterprise processing is governed separately.
  deletion_controls: Privacy rights and account deletion requests exist, subject
    to legal and operational exceptions; no per-request API deletion control is
    published.
  caveat: Sarvam's public privacy policy excludes enterprise-customer content from
    its scope, but free developer access should not be assumed to receive
    enterprise protections.
agreements:
  terms_of_service: https://www.sarvam.ai/terms-of-service
  privacy_policy: https://www.sarvam.ai/privacy-policy
```

## Eligibility

```yaml
account_required: true
payment_method_required: not_explicitly_stated
```

## API compatibility and modalities

- **Compatibility:** Sarvam Rest With OpenAI Style Chat Parameters
- **Modalities:** Text generation
- **Geography:** India

## Before you build with Sarvam AI

### Read the classification narrowly

This record describes the bounded offer supported by evidence on 2026-09-19; it does not guarantee permanence, production suitability, uptime, latency, model quality, or access from every account and region. “One-time trial” should be interpreted together with the Current directory placement, High confidence, No payment-card status, and the plan-specific sources below.

### Resolve the live model route

The snapshot records 5 model IDs. Match the exact ID against the provider's current catalog before using it in code because zero-price routes, aliases, context limits, and feature support can rotate while an older documentation page remains online.

### Match the quota to the workload shape

Translate the allowance into peak requests per minute, input and output tokens, concurrency, retries, and every daily or monthly ceiling that applies to the intended account. The first limit reached by the workload is the practical ceiling. A large token pool can still fail interactive bursts, and a high request limit can still fail long-context work. Include tool calls and retry traffic, test the largest realistic payload, and treat research, experimental, and community access as conditional on their eligibility and fair-use terms.

### Preserve billing and privacy boundaries

The recorded payment-card requirement is No, and account access is required. Confirm both in the live signup flow, set provider-side budgets when charges are possible, and observe the balance or usage fields after a complete request. The prompt-handling classification is **Not private** because Free developer inputs and outputs are collected and may support service improvement without a clear default no-training commitment. Re-read the terms for the exact route and plan before sending sensitive, regulated, or proprietary content.

### Follow the evidence, then re-check it

This record links 10 first-party sources covering the offer, catalog, limits, pricing, terms, privacy, or adoption evidence available to the audit. Prefer the newest and most specific governing document or live catalog when sources disagree. A dated finding can become stale even when the page remains online, so retain the source and snapshot that supported the decision and submit a correction when a provider changes a material term.

### Plan fallback without policy drift

A fallback should preserve modality, context length, streaming, structured output, tools, safety controls, and data terms, not only API syntax. Decide which errors may retry, cap retry storms, and prevent an exhausted free route from silently switching to a billable model. Rotating aliases and free pools can change behavior without changing the endpoint, so retain the response model field and test at least one substitute route before the primary offer becomes unavailable.

### Monitor the offer as a dependency

Capture the model ID, response model, rate-limit headers, usage fields, latency, HTTP status, and provider request identifier. Watch authorization failure, quota exhaustion, catalog removal, policy revision, and balance movement as separate failure modes. Re-check the live catalog and governing sources on a schedule proportionate to the workload's importance, and keep an owner and exit path for any production dependency on volatile free capacity.

### Test one complete request before scaling

Start with the smallest permitted request using the exact credential, model ID, endpoint, region, and account type intended for deployment. Record the status, headers, usage fields, response model, latency, and dashboard balance movement. Then exercise an invalid model, quota exhaustion, or rate limit so failure is explicit and cannot silently switch to a paid route. Validate streaming, structured output, and tool calls separately because a free model can expose fewer features than its paid counterpart. Keep a budget ceiling outside the application whenever billing is possible, and do not send sensitive data until the observed route matches the reviewed agreement.

## Primary sources

- [Rate limits](https://docs.sarvam.ai/api/getting-started/ratelimits): Official Docs
- [Open-source models](https://docs.sarvam.ai/api/getting-started/models/open-source): Official Catalog
- [API pricing](https://web.sarvam.dev/api-pricing): Official Pricing
- [Models API](https://api.sarvam.ai/v2/models): Live Catalog
- [Sarvam Terms of Service](https://www.sarvam.ai/terms-of-service): Official Terms
- [Sarvam Privacy Policy](https://www.sarvam.ai/privacy-policy): Official Privacy
- [Sarvam Trust Center](https://www.sarvam.ai/trust-center): Official Security
- [Inc42: Sarvam to build trillion-parameter model, developer platform passes 1 Mn registered developers (2026-07-30)](https://inc42.com/buzz/sarvam-to-build-trillion-plus-ai-model-in-india-launches-inference-service/): Reputable Reporting
- [PyPI download stats for sarvamai (official Sarvam Python SDK)](https://pypistats.org/packages/sarvamai): Package Registry Stats
- [Sarvam AI organization on Hugging Face (Sarvam first-party model family)](https://huggingface.co/sarvamai): Official Repository

## Sitemap

See the full [semantic sitemap](/sitemap.md) for every page and markdown mirror.