Sarvam AI: ₹1,000 free API credits (22 Indian languages STT/TTS/translation)
Source: https://www.sarvam.ai/api-pricing
Description
Create account to comment on specific lines or Sign in
+ 1 Sarvam AI is India's sovereign full-stack AI platform specializing in 22 officially recognized Indian languages. Every new developer account gets ₹1,000 in free credits (per the public pricing page) usable across speech-to-text, text-to-speech, translation, transliteration, chat completion, vision/document intelligence, and language identification APIs. Credits never expire and are universal across every Sarvam API. No upfront payment is required to start — the Starter plan is pure pay-as-you-go.
No comments on this line yet.
+ 2
No comments on this line yet.
+
3
Heads up — credit amount discrepancy: The public pricing page advertises ₹1,000 free credits for every new signup, but the official rate-limits docs currently say ₹100 complimentary credits. The marketing page is more recent; either way you get something for free without entering a card. Treat ₹100 as the worst-case floor.
Heads up — credit amount discrepancy: The public pricing page advertises ₹1,000 free credits for every new signup, but the official rate-limits docs currently say ₹100 complimentary credits. The marketing page is more recent; either way you get something for free without entering a card. Treat ₹100 as the worst-case floor.
No comments on this line yet.
+ 4
No comments on this line yet.
+
5
No comments on this line yet.
+ 6
No comments on this line yet.
+ 8
No comments on this line yet.
+ 10 2. Sign up with your email (Google sign-in also typically available on Indian dev platforms)
No comments on this line yet.
+ 11 3. Verify your email
No comments on this line yet.
+ 12 4. Complete a short profile (name, intended use case)
No comments on this line yet.
+ 13 5. Free credits are automatically provisioned to your account on first login — no credit card requested for the Starter plan
No comments on this line yet.
+ 14 6. Generate your API key from the dashboard
No comments on this line yet.
+ 15 7. Start calling the REST endpoints directly or use the OpenAI-compatible chat-completions endpoint for LLMs
No comments on this line yet.
+ 16
No comments on this line yet.
+ 17 Important caveats:
No comments on this line yet.
+ 18 • No credit card required to use the free credits or stay on the Starter (pay-as-you-go) plan
No comments on this line yet.
+ 19 • Credits and rate-limit tier are decoupled — buying more credits does NOT increase your RPM. To get higher rate limits you must upgrade plans (Pro / Business / Enterprise)
No comments on this line yet.
+ 20 • Account-level rate limits are shared across all your API keys (you cannot bypass limits by minting more keys)
No comments on this line yet.
+ 21 • For volume beyond Business tier, email developer@sarvam.ai for Enterprise pricing
No comments on this line yet.
+ 22 • All 22 Indian languages are accessible from day one — no separate approval needed
No comments on this line yet.
+ 23
No comments on this line yet.
+
24
No comments on this line yet.
+ 25
No comments on this line yet.
+ 27
No comments on this line yet.
+ 29 ModelNotes
No comments on this line yet.
+ 30 Sarvam-M (24B)Open-weight, hybrid reasoning, Indic-tuned. Free to call via API
No comments on this line yet.
+ 31 Sarvam-105BFrontier Indic LLM (subject to capacity, free per pricing page)
No comments on this line yet.
+ 32 Sarvam-30BMid-size Indic LLM, free per pricing page
No comments on this line yet.
+ 33
No comments on this line yet.
+ 35 • Saaras v2 / v2.5 — supports all 22 scheduled Indian languages, 8 kHz telephony audio, code-mixed speech
No comments on this line yet.
+ 36 • Variants: standard transcription, with diarization, and with translation-to-English
No comments on this line yet.
+ 37
No comments on this line yet.
+ 39 • Bulbul v3 (latest) — natural multi-speaker TTS
No comments on this line yet.
+ 40 • Bulbul v2 — earlier version, still supported
No comments on this line yet.
+ 41 • Available in 11 Indian languages with 39 distinct speaker voices
No comments on this line yet.
+ 42
No comments on this line yet.
+ 44 • Sarvam-Translate v1 — paragraph-level translation across all 22 languages, structured content for 15 of them (built on Gemma 3 with DeepMind collaboration)
No comments on this line yet.
+ 45 • Mayura v1 — translation with explicit script control (e.g., force Devanagari output)
No comments on this line yet.
+ 46
No comments on this line yet.
+ 48 • Transliteration — convert text between Indic scripts and Roman
No comments on this line yet.
+ 49 • Language Identification — detect which Indic language a snippet is in
No comments on this line yet.
+ 50 • Vision / Document Intelligence — image understanding & document digitization (₹1.50 per page)
No comments on this line yet.
+ 51
No comments on this line yet.
+ 52 Source: Sarvam Models Overview and Models Documentation. Verified 2026-05-09.
No comments on this line yet.
+ 53
No comments on this line yet.
+
54
No comments on this line yet.
+ 55
No comments on this line yet.
+ 57
No comments on this line yet.
+ 58 Full list (Sarvam-Translate / Saaras STT):
No comments on this line yet.
+ 59
No comments on this line yet.
+ 60 Hindi, Bengali, Marathi, Telugu, Tamil, Gujarati, Urdu, Kannada, Odia, Malayalam, Punjabi, Assamese, Maithili, Santali, Kashmiri, Nepali, Sindhi, Dogri, Konkani, Manipuri (Meitei), Bodo, and Sanskrit.
No comments on this line yet.
+ 61
No comments on this line yet.
+ 62 Note: Bulbul TTS currently covers a subset of these (11 languages, 39 voices).
No comments on this line yet.
+ 63
No comments on this line yet.
+
64
No comments on this line yet.
+ 65
No comments on this line yet.
+ 67
No comments on this line yet.
+ 68 ServiceApprox. Rate
No comments on this line yet.
+ 69 Chat LLMs (Sarvam-M / 30B / 105B)Free
No comments on this line yet.
+ 70 Speech-to-Text₹30–45 per hour of audio
No comments on this line yet.
+ 71 Text-to-Speech (Bulbul)₹15–30 per 10,000 characters
No comments on this line yet.
+ 72 Translation / Transliteration₹20 per 10,000 characters
No comments on this line yet.
+ 73 Vision / Document Intelligence₹1.50 per page
No comments on this line yet.
+ 74
No comments on this line yet.
+ 75 At these rates, ₹1,000 in credits realistically buys roughly 22+ hours of STT, or ~330,000 characters of TTS, or ~500K characters of translation — enough for serious prototyping.
No comments on this line yet.
+ 76
No comments on this line yet.
+
77
No comments on this line yet.
+ 78
No comments on this line yet.
+ 80
No comments on this line yet.
+ 81 PlanCostBonus CreditsRate LimitSupport
No comments on this line yet.
+ 82 Starter₹0 (PAYG)₹1,000 signup60 RPMCommunity
No comments on this line yet.
+ 83 Pro₹10,000₹1,000 bonus (₹11,000 total)200 RPMStandard
No comments on this line yet.
+ 84 Business₹50,000₹7,500 bonus (₹57,500 total)1,000 RPMSlack support
No comments on this line yet.
+ 85 EnterpriseContactCustomCustomDedicated
No comments on this line yet.
+ 86
No comments on this line yet.
+ 87 Rate limits operate in three concurrency modes (Provisioned, Burst, High Throughput) and are enforced per account.
No comments on this line yet.
+ 88
No comments on this line yet.
+
89
No comments on this line yet.
+ 90
No comments on this line yet.
+ 92
No comments on this line yet.
+ 93 • No catch on the free tier — you genuinely don't need a card. Credits don't expire.
No comments on this line yet.
+ 94 • The Starter plan caps you at 60 RPM, which is fine for prototyping but you'll feel it on production loads. Going higher RPM requires upgrading (you can't just buy more credits).
No comments on this line yet.
+ 95 • Documentation is bilingual but improving — some pages (e.g. registration flow) are sparse. Use the dashboard UX rather than the docs for the signup walkthrough.
No comments on this line yet.
+ 96 • Geographic focus: the platform is engineered for Indic languages. Don't expect it to be competitive on European or Mandarin TTS — that's not the product.
No comments on this line yet.
+ 97 • The ₹1,000 vs ₹100 documentation conflict (see top of page) means you should verify what actually shows up in your dashboard after signup, and screenshot it for your records.
No comments on this line yet.
+ 98
No comments on this line yet.
+
99
No comments on this line yet.
+ 100
No comments on this line yet.
+ 102
No comments on this line yet.
+ 103 • Sarvam Startup Programme (launched 5 March 2026) — selected early-stage Indian AI startups get 6–12 months of API credits plus engineering support and co-marketing. Apply via sarvam.ai if you're an Indian-incorporated AI startup. (Listed separately as Entry 52.)
No comments on this line yet.
+ 104 • IndiaAI Mission compute pool (~40,000 GPUs) — complementary government program for Indian deep-tech startups.
No comments on this line yet.
+ 105
No comments on this line yet.
+
106
No comments on this line yet.
+ 107
No comments on this line yet.
+ 109
No comments on this line yet.
+
110
• OpenAI-compatible endpoint is available for chat — drop-in replacement for openai.ChatCompletion calls if you want to test Sarvam-M against GPT/Claude in the same harness
No comments on this line yet.
+ 111 • The chat LLMs are listed as free on the pricing page, so you can run unlimited Sarvam-M calls without burning your ₹1,000 credit pool — save the credits for STT/TTS/translation
No comments on this line yet.
+ 112 • Code-mixed support (e.g., Hinglish) is a real differentiator — Saaras handles it natively, where Western STT models struggle
No comments on this line yet.
+ 113 • 8 kHz telephony audio is supported out of the box on Saaras — useful for call-center transcription
No comments on this line yet.
+ 114 • For voice-first agentic apps in Indian languages, the Bulbul TTS + Saaras STT + Sarvam-M LLM combo is the easiest end-to-end Indic stack
No comments on this line yet.
+ 115
No comments on this line yet.
+
116
No comments on this line yet.
+ 117
No comments on this line yet.
+ 118 Sources:
No comments on this line yet.
+ 119 • Sarvam AI API Pricing
No comments on this line yet.
+ 120 • Sarvam AI Credits & Rate Limits Docs
No comments on this line yet.
+ 121 • Sarvam Models Overview
No comments on this line yet.
+ 122 • Sarvam-Translate Model Docs
No comments on this line yet.
+ 123 • Sarvam-Translate Blog Post
No comments on this line yet.
+ 124 • Sarvam AI Wikipedia
No comments on this line yet.
+ 125 • Google DeepMind: Sarvam AI + Gemma 3 for 22 Indian languages
No comments on this line yet.
+ 126 • Sarvam AI Startup Programme (BusinessToday, 2026-03-05)
No comments on this line yet.