Languages with named standard API voices in the reviewed voice table.
Context caching
Not collected
The public gRPC request has rebuild_cache, but cache semantics, eligibility, TTL and billing are not documented by that field. No free-cache claim is made.
Endpoint availability
New business customers cannot purchase API packages or start pay-as-you-go from 15 July 2026. Existing eligible business customers retain documented tariffs. Personal package sales and Freemium renewals also stopped on that date; remaining packages expire normally.Source for Endpoint availability (opens in a new tab)
Synchronous HTTP accepts up to 4,000 characters. Asynchronous synthesis accepts a text file without SSML. The limits page lists a 1 GB ceiling for streaming.Source for Input requirements (opens in a new tab)
Language parameter options in the gRPC reference. Matching public voice IDs for every language were not established; do not assume all named voices speak every language.
Existing eligible business customers only; new purchases closed 15 July 2026. RUB including VAT. Pay-as-you-go has a RUB 15,000 monthly minimum when usage is nonzero; no charge for an unused month. Usage amounts below are proportional charges before the monthly minimum, not a standalone subscription quote. Billed characters include spaces and SSML markup.Source for Billing basis (opens in a new tab)
Access & pricing
Prices belong to an endpoint and billing tier. A provider listing does not guarantee access for every account or region.
Prices retain their published currencies and billing units below.
Published billing units · original currency · observed 14 Oct 2026
Category
Usage
Rate
Unit
Existing business pay-as-you-go
Speech synthesis characters1
RUB 0.000186
character
Rate conditions
Existing eligible business customers only; new purchases closed 15 July 2026. RUB including VAT. Pay-as-you-go has a RUB 15,000 monthly minimum when usage is nonzero; no charge for an unused month. Usage amounts below are proportional charges before the monthly minimum, not a standalone subscription quote. Counts spaces and SSML markup, where SSML is supported.
Prices retain their published currencies and billing units below.
Published billing units · original currency · observed 14 Oct 2026
Category
Usage
Rate
Unit
Existing business pay-as-you-go
Speech synthesis characters1
RUB 0.000186
character
Rate conditions
Existing eligible business customers only; new purchases closed 15 July 2026. RUB including VAT. Pay-as-you-go has a RUB 15,000 monthly minimum when usage is nonzero; no charge for an unused month. Usage amounts below are proportional charges before the monthly minimum, not a standalone subscription quote. Counts spaces and SSML markup, where SSML is supported.
Prices retain their published currencies and billing units below.
Published billing units · original currency · observed 14 Oct 2026
Category
Usage
Rate
Unit
Existing business pay-as-you-go
Speech synthesis characters1
RUB 0.000186
character
Rate conditions
Existing eligible business customers only; new purchases closed 15 July 2026. RUB including VAT. Pay-as-you-go has a RUB 15,000 monthly minimum when usage is nonzero; no charge for an unused month. Usage amounts below are proportional charges before the monthly minimum, not a standalone subscription quote. Counts spaces and SSML markup, where SSML is supported.
Audio retention, model-training use and account-specific residency terms not established in this review.
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 INPUTProcess & retain
The system processes eligible input and retains reusable state.
02 / LATER REQUESTCheck available state
Reuse depends on matching input, session state and the cache policy.
03 / CACHE HITReuse & generate
Reusing state can reduce repeated work. Any charges follow the billing rules below.
Sber SaluteSpeech
Not verified
No verified model-specific cache TTL, discount or hit-accounting policy. GigaChat text caching does not apply by inference.
How it works
Not collected
No verified model-specific cache TTL, discount or hit-accounting policy. GigaChat text caching does not apply by inference.
Minimum reusable input
Not collected
No verified model-specific cache TTL, discount or hit-accounting policy. GigaChat text caching does not apply by inference.
Lifetime & refresh
Not collected
No verified model-specific cache TTL, discount or hit-accounting policy. GigaChat text caching does not apply by inference.
How to enable it
Not collected
No verified model-specific cache TTL, discount or hit-accounting policy. GigaChat text caching does not apply by inference.
Usage fields
Not collected
No verified model-specific cache TTL, discount or hit-accounting policy. GigaChat text caching does not apply by inference.
Billing rules
Not collected
No verified model-specific cache TTL, discount or hit-accounting policy. GigaChat text caching does not apply by inference.
Limits & data boundaries
Not collected
No verified model-specific cache TTL, discount or hit-accounting policy. GigaChat text caching does not apply by inference.
No verified cache rate for this endpoint. An empty rate is not a zero charge.
Caching reviewed 14 Oct 2026
Sber SaluteSpeech
Not verified
No verified model-specific cache TTL, discount or hit-accounting policy. GigaChat text caching does not apply by inference.
How it works
Not collected
No verified model-specific cache TTL, discount or hit-accounting policy. GigaChat text caching does not apply by inference.
Minimum reusable input
Not collected
No verified model-specific cache TTL, discount or hit-accounting policy. GigaChat text caching does not apply by inference.
Lifetime & refresh
Not collected
No verified model-specific cache TTL, discount or hit-accounting policy. GigaChat text caching does not apply by inference.
How to enable it
Not collected
No verified model-specific cache TTL, discount or hit-accounting policy. GigaChat text caching does not apply by inference.
Usage fields
Not collected
No verified model-specific cache TTL, discount or hit-accounting policy. GigaChat text caching does not apply by inference.
Billing rules
Not collected
No verified model-specific cache TTL, discount or hit-accounting policy. GigaChat text caching does not apply by inference.
Limits & data boundaries
Not collected
No verified model-specific cache TTL, discount or hit-accounting policy. GigaChat text caching does not apply by inference.
No verified cache rate for this endpoint. An empty rate is not a zero charge.
Caching reviewed 14 Oct 2026
Sber SaluteSpeech
Not verified
No verified model-specific cache TTL, discount or hit-accounting policy. GigaChat text caching does not apply by inference.
How it works
Not collected
No verified model-specific cache TTL, discount or hit-accounting policy. GigaChat text caching does not apply by inference.
Minimum reusable input
Not collected
No verified model-specific cache TTL, discount or hit-accounting policy. GigaChat text caching does not apply by inference.
Lifetime & refresh
Not collected
No verified model-specific cache TTL, discount or hit-accounting policy. GigaChat text caching does not apply by inference.
How to enable it
Not collected
No verified model-specific cache TTL, discount or hit-accounting policy. GigaChat text caching does not apply by inference.
Usage fields
Not collected
No verified model-specific cache TTL, discount or hit-accounting policy. GigaChat text caching does not apply by inference.
Billing rules
Not collected
No verified model-specific cache TTL, discount or hit-accounting policy. GigaChat text caching does not apply by inference.
Limits & data boundaries
Not collected
No verified model-specific cache TTL, discount or hit-accounting policy. GigaChat text caching does not apply by inference.
No verified cache rate for this endpoint. An empty rate is not a zero charge.
Caching reviewed 14 Oct 2026
Voice choices
7 voices
Choose a voice within SaluteSpeech Text-to-Speech. These are API voice identifiers, not additional model releases. Source links document each voice.
Published voice catalog
Voice / API ID
Language & accent
Delivery
NataliaNec_24000
Russian
24 kHz voice. Use Nec_8000 for the 8 kHz telephony variant. These sample-rate variants are the same named voice, not separate foundation models.Voice source (opens in a new tab)
BorisBys_24000
Russian
24 kHz voice. Use Bys_8000 for the 8 kHz telephony variant. These sample-rate variants are the same named voice, not separate foundation models.Voice source (opens in a new tab)
MarfaMay_24000
Russian
24 kHz voice. Use May_8000 for the 8 kHz telephony variant. These sample-rate variants are the same named voice, not separate foundation models.Voice source (opens in a new tab)
TarasTur_24000
Russian
24 kHz voice. Use Tur_8000 for the 8 kHz telephony variant. These sample-rate variants are the same named voice, not separate foundation models.Voice source (opens in a new tab)
AlexandraOst_24000
Russian
24 kHz voice. Use Ost_8000 for the 8 kHz telephony variant. These sample-rate variants are the same named voice, not separate foundation models.Voice source (opens in a new tab)
SergeyPon_24000
Russian
24 kHz voice. Use Pon_8000 for the 8 kHz telephony variant. These sample-rate variants are the same named voice, not separate foundation models.Voice source (opens in a new tab)
KiraKin_24000
English
24 kHz voice. Use Kin_8000 for the 8 kHz telephony variant. These sample-rate variants are the same named voice, not separate foundation models.Voice source (opens in a new tab)
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
Provider-documented hosted speech component. Public API schemas do not publish the served weights. The exact underlying checkpoint and its relationship to public GigaAM releases are not established, so this record does not assert a separate proprietary foundation model.
