Pricing
Transparent pricing for all Sarvam AI services. All prices are in Indian Rupees (₹).
Chat Completion
₹29.28 / ₹10.98 / ₹73.2 per 1M tokens
Chat completion. Input ₹29.28 · Cached input ₹10.98 · Output ₹73.2 — per 1M tokens.
₹29.28 / ₹10.98 / ₹73.2 per 1M tokens
Chat completion (sarvam-105b-conversations). Input ₹29.28 · Cached input ₹10.98 · Output ₹73.2 — per 1M tokens.
₹36.6 / ₹13.73 / ₹91.5 per 1M tokens
Chat completion (beta). Input ₹36.6 · Cached input ₹13.73 · Output ₹91.5 — per 1M tokens.
₹128.1 / ₹23.79 / ₹402.6 per 1M tokens
Chat completion (beta). Input ₹128.1 · Cached input ₹23.79 · Output ₹402.6 — per 1M tokens.
GLM-5.2 and Gemma 4 31B are available in beta and rolling out gradually. GLM-5.2 reasons before every answer, and reasoning tokens are billed as output tokens — see Reasoning for how to budget max_tokens.
Speech
Billed per second of audio
₹30/hour
Charged per second of audio. Rounded up to the nearest second in each request.
₹45/hour
Adds speaker identification. Rounded up to the nearest second in each request.
₹30/hour
Transcribes and translates. Rounded up to the nearest second in each request.
₹45/hour
Transcribes, translates, and identifies speakers. Rounded up to the nearest second in each request.
Language Tools
Billed per character
₹20/10K characters
Rounded up to the nearest character in each request.
₹20/10K characters
Rounded up to the nearest character in each request.
₹20/10K characters
Rounded up to the nearest character in each request.
₹3.5/10K characters
Rounded up to the nearest character in each request.
Text to Speech
₹15/10K characters
Charged per character. Rounded up to the nearest character in each request.
₹30/10K characters
Beta pricing. Charged per character. Rounded up to the nearest character in each request.
Document Digitization
₹0.5/page
Extract structured data, tables, and text from documents with our powerful document parsing API. Maximum 10 pages per job.
Dubbing
Billed per second of source media, multiplied by the number of target languages
₹40/min (Starter)
Default for API integrations with editor_flow: false. Charged per whole second of source media × each target language.
₹80/min (Starter)
Applies when editor_flow: true. Costs twice as many credits as the standard API rate and enables the interactive editor workflow in Creator Studio.
Keep editor_flow at false for standard API integrations. Setting it to true switches billing to the editor-flow rate (₹80/min on Starter — double the default API rate) and suppresses auto-export. Only enable it when you deliberately need the editor workflow.
Duration is truncated to whole seconds before billing. A 90-second file dubbed into Hindi and Tamil on Starter at the default API rate costs roughly ₹120 (90s × 2 languages × ₹40/60 per second). See the Dubbing overview for integration details.
Pricing Summary
Rate Limits by Plan
Each plan includes account-level rate limits that apply across all APIs:
For full details including per-API limits, HTTP response headers, and error handling, see the Credits & Rate Limits page.
Enterprise & Custom Pricing
Need custom rate limits, volume discounts, or dedicated support? Contact our team for enterprise pricing.
Get technical support, integration help, and custom solutions from our developer team.
Free Credits: Every new user receives ₹100 worth of free credits to explore all our APIs. Check our Credits & Rate Limits page for more details. For commercial production rights when shipping generated audio (including Bulbul v3 and voice cloning), see Commercial Licensing.