/ Sarvam AI / Bulbul v4 Flash Sarvam AI / Bulbul
Bulbul v4 Flash Speech synthesis with 222 personas organized by language and delivery style.
available Closed weights Text-to-speech
Text input limit 2,500 characters per REST request See specification notes below
Languages 10 listed Full language list in specifications
Input billing Not collected Sarvam API
Output billing Not collected No inferred price
Specifications Access & pricing Caching Voices Evaluation Sources & lifecycle
Specifications Provider-documented facts. “Not collected” means we have not verified that field yet, not that the feature is unsupported.
Context / input limit Not collected
Maximum output Not collected
Reasoning Not collected
Tool calling Not collected
Structured output Not collected
Batch processing Not collected
Fine-tuning Not collected
Knowledge cutoff Not collected
Release date Not collected
Parameters Not collected
Architecture Not collected Languages Hindi, English, English–Hindi code-mixed, Bengali, Gujarati, Kannada, Marathi, Punjabi, Tamil, TeluguSource for Languages (opens in a new tab) Voice-catalog coverage, not the larger v3 language-code list. Malayalam and Odia personas are not yet listed for v4 Flash.
Access & pricing Prices belong to an endpoint and billing tier. A provider listing does not guarantee access for every account or region.
API model ID bulbul:v4-flash
Access Sarvam API subscription key; credit balance and account rate limits apply.
Regions Model-specific regional routing and contractual data residency require verification. Endpoint pricing has not been collected. Check the provider for applicable rates.
Data & enterprise terms Production rights apply to output generated while account credits are consumed, including signup credits, subject to EULA terms. Rights to the underlying model and voices are not transferred.
Read the service policy (opens in a new tab)
API model ID bulbul:v4-flash
Access Sarvam API subscription key; credit balance and account rate limits apply.
Regions Model-specific regional routing and contractual data residency require verification. Endpoint pricing has not been collected. Check the provider for applicable rates.
Data & enterprise terms Retention, training use, regional processing and enterprise terms have not been summarized for this endpoint.
API model ID bulbul:v4-flash
Access Sarvam API subscription key; credit balance and account rate limits apply.
Regions Model-specific regional routing and contractual data residency require verification. Endpoint pricing has not been collected. Check the provider for applicable rates.
Data & enterprise terms Retention, training use, regional processing and enterprise terms have not been summarized for this endpoint.
Caching Caching can reuse earlier input processing. Availability, lifetime and billing depend on the endpoint. A cache hit still generates a new response.
01 / INITIAL INPUT Process & retain The system processes eligible input and retains reusable state.
02 / LATER REQUEST Check available state Reuse depends on matching input, session state and the cache policy.
03 / CACHE HIT Reuse & generate Reusing state can reduce repeated work. Any charges follow the billing rules below.
Sarvam API bulbul:v4-flashNot supported Beta response-audio caching is documented only for v1/v2, not v3/v4 Flash. This is not token-prefix caching.
Support evidence (opens in a new tab) Caching reviewed 13 Oct 2026
Sarvam API bulbul:v4-flashNot verified
How it works Not collected
Minimum reusable input Not collected
Lifetime & refresh Not collected
How to enable it Not collected
Usage fields Not collected
Billing rules Not collected
Limits & data boundaries Not collected No verified cache rate for this endpoint. An empty rate is not a zero charge.
Sarvam API bulbul:v4-flashNot verified
How it works Not collected
Minimum reusable input Not collected
Lifetime & refresh Not collected
How to enable it Not collected
Usage fields Not collected
Billing rules Not collected
Limits & data boundaries Not collected No verified cache rate for this endpoint. An empty rate is not a zero charge.
What the measurements say No benchmark observations have been added for this model. We do not turn a vendor description into a performance score.
Latency and throughput depend on the host, request size, reasoning settings and load. There is no measured speed or global usage figure in this record.
The record behind the model Why it is included Inferred for this exact native API release/service from provider documentation and review of Sarvam’s public weight inventory. No weights for this identity were identified. Public Sarvam language/translation weights and hosted third-party weights are excluded. A named service or version does not establish an independently disclosed checkpoint.
Classification evidence (opens in a new tab) Sources reviewed Model notes 14 reviewed release and service records, including 222 Bulbul v4 Flash personas. Public-weight language models are excluded. Historical versions, ambiguous conversational-model licensing and some deployment details remain under review.
Named voices are presets, not separate model releases. REST defaults to v3, while the streaming references still specify v2 as default. Pin the model and choose a compatible speaker explicitly.
The current price table names Bulbul v3 only. Its ₹30/10K-character rate is not automatically assigned to this version.
Check against your workload Try representative inputs, measure task success and total cost, then check the exact endpoint’s data policy.
Read our methodology
Also from Sarvam AI Bulbul v1 Bulbul v2 Bulbul v3 Saaras v2.5 Saaras v3 Saaras v3-realtime Saaras v4 Saarika v2.5 Mayura v1 Instant Voice Cloning Sarvam Vision Sarvam Vision 1.5 