Skip to content
Model infoModelsSnapshot: 15 Oct 2026

Smallest AI / Lightning

Lightning V3.1

Streaming multilingual speech synthesis with standard and cloned voices, pronunciation controls and 44.1 kHz native audio.

Speech & voice
availableClosed weightsText-to-speech
Official documentation (opens in a new tab)Reviewed 13 Oct 2026
12 of 24 specification fields collected
Output formatsPCM, MP3, WAV, ulaw, alawSee specification notes below
Languages20 listedFull language list in specifications
Input billingSee ratesSmallest AI API
Output billingSee ratesPublished units and conditions below

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
Input modalities
TextSource for Input modalities (opens in a new tab)
Output modalities
AudioSource for Output modalities (opens in a new tab)
Reasoning
Not collected
Tool calling
Not collected
Structured output
Not collected
Streaming
SupportedSource for Streaming (opens in a new tab)
Batch processing
Not collected
Fine-tuning
Not collected
Knowledge cutoff
Not collected
Release date
Not collected
Parameters
Not collected
Architecture
Not collected
Languages
View all 20 entries
  • English
  • Hindi
  • Tamil
  • Spanish
  • Kannada
  • Marathi
  • Telugu
  • Odia
  • Punjabi
  • Malayalam
  • Gujarati
  • Bangla
  • French
  • German
  • Italian
  • Dutch
  • Swedish
  • Portuguese
  • Polish
  • Russian
Source for Languages (opens in a new tab)

20 accepted codes; only 12 have trained voice catalogs. Eight extra codes route through English/Hindi voices.

Context caching
Not collected
Native model options
lightning-v3.1Source for Native model options (opens in a new tab)

Registry identifier uses hyphens; unified TTS request selectors for v3.1/v3.2 use underscores.

Voice cloning
SupportedSource for Voice cloning (opens in a new tab)

Voice cloning is not evidence of user access to base-model fine-tuning. Legacy flags do not imply current clone creation works.

Stock voice catalog
234 public stock voices mapped to lightning-v3.1 on 2026-10-13Source for Stock voice catalog (opens in a new tab)

Snapshot count, not a permanent service limit. Presets are not independent model checkpoints.

Audio sample rates
8,000 / 16,000 / 24,000 / 44,100 Hz; native 44.1 kHzSource for Audio sample rates (opens in a new tab)
Output formats
PCM, MP3, WAV, ulaw, alawSource for Output formats (opens in a new tab)
Streaming audio chunks
Up to 250 characters per chunk; about 140 recommended for throughputSource for Streaming audio chunks (opens in a new tab)

Chunk limit, not the total request or context length.

latency
About 200 ms time to first byte at 40 concurrent requests, measured in-regionSource for latency (opens in a new tab)

Provider measurement, not an end-to-end SLA. Marketing and streaming pages use other conditions.

Input requirements
Instant cloning: clean 5–15 second sample under 5 MB; MP3, WAV, MP4 or WebMSource for Input requirements (opens in a new tab)

Optional clone-enrollment input, not the text synthesis input limit. Clones must match their target model pool.

Access & pricing

Prices belong to an endpoint and billing tier. A provider listing does not guarantee access for every account or region.

Smallest AI API

Provider source (opens in a new tab)
API model ID
lightning_v3.1
Access
Synchronous POST at https://api.smallest.ai/waves/v1/tts. API key and matching voice_id required.
Regions
Automatic routing to Mumbai, India or Oregon, USA; api.india.smallest.ai and api.us.smallest.ai pin a region. Language availability can differ by region.
API endpoint URL
https://api.smallest.ai/waves/v1/ttsSource for API endpoint URL (opens in a new tab)
Billing basis
Characters synthesized; enterprise rates by agreementSource for Billing basis (opens in a new tab)
Request quotas
Pricing page: 15 TTS streams / 100 RPM; generic limits guide: 1 active request shared across models. Conflicting published defaults.Source for Request quotas (opens in a new tab)
Word timestamps
Not supportedSource for Word timestamps (opens in a new tab)

Model-card contract limits timing events to supported English/Hindi voices over WebSocket. HTTP/SSE accept but ignore the flag; verify the selected voice.

Prices retain their published currencies and billing units below.

Published billing units · original currency · observed 13 Oct 2026
CategoryUsageRateUnit
Speechpayg character rate1$0.17510,000 characters

Rate conditions

  1. Approximate published pay-as-you-go rate. Enterprise pricing is custom. This is character billing, not text-token billing.
Pricing source (opens in a new tab)

Data & enterprise terms

Content is retained unless an eligible Enterprise data policy is enabled. Options: no content storage, a 24-hour to 90-day window, or x-expire-content: true for seven-day expiry. On other plans the header is ignored. Billing/usage metadata remains; settings are not retroactive. TTS saved history is excluded from window expiry. Speech-to-speech and cloning are not expressly covered.

Read the service policy (opens in a new tab)

Smallest AI API

Provider source (opens in a new tab)
API model ID
lightning_v3.1
Access
POST with server-sent events at https://api.smallest.ai/waves/v1/tts/live. API key and matching voice_id required.
Regions
Automatic routing to Mumbai, India or Oregon, USA; api.india.smallest.ai and api.us.smallest.ai pin a region. Language availability can differ by region.
API endpoint URL
https://api.smallest.ai/waves/v1/tts/liveSource for API endpoint URL (opens in a new tab)
Billing basis
Characters synthesized; enterprise rates by agreementSource for Billing basis (opens in a new tab)
Request quotas
Pricing page: 15 TTS streams / 100 RPM; generic limits guide: 1 active request shared across models. Conflicting published defaults.Source for Request quotas (opens in a new tab)
Word timestamps
Not supportedSource for Word timestamps (opens in a new tab)

Model-card contract limits timing events to supported English/Hindi voices over WebSocket. HTTP/SSE accept but ignore the flag; verify the selected voice.

Prices retain their published currencies and billing units below.

Published billing units · original currency · observed 13 Oct 2026
CategoryUsageRateUnit
Speechpayg character rate1$0.17510,000 characters

Rate conditions

  1. Approximate published pay-as-you-go rate. Enterprise pricing is custom. This is character billing, not text-token billing.
Pricing source (opens in a new tab)

Data & enterprise terms

Content is retained unless an eligible Enterprise data policy is enabled. Options: no content storage, a 24-hour to 90-day window, or x-expire-content: true for seven-day expiry. On other plans the header is ignored. Billing/usage metadata remains; settings are not retroactive. TTS saved history is excluded from window expiry. Speech-to-speech and cloning are not expressly covered.

Read the service policy (opens in a new tab)

Smallest AI API

Provider source (opens in a new tab)
API model ID
lightning_v3.1
Access
Bidirectional WebSocket at wss://api.smallest.ai/waves/v1/tts/live. API key and matching voice_id required.
Regions
Automatic routing to Mumbai, India or Oregon, USA; api.india.smallest.ai and api.us.smallest.ai pin a region. Language availability can differ by region.
API endpoint URL
wss://api.smallest.ai/waves/v1/tts/liveSource for API endpoint URL (opens in a new tab)
Billing basis
Characters synthesized; enterprise rates by agreementSource for Billing basis (opens in a new tab)
Request quotas
Pricing page: 15 TTS streams / 100 RPM; generic limits guide: 1 active request shared across models. Conflicting published defaults.Source for Request quotas (opens in a new tab)
Word timestamps
SupportedSource for Word timestamps (opens in a new tab)

Model-card contract limits timing events to supported English/Hindi voices over WebSocket. HTTP/SSE accept but ignore the flag; verify the selected voice.

Prices retain their published currencies and billing units below.

Published billing units · original currency · observed 13 Oct 2026
CategoryUsageRateUnit
Speechpayg character rate1$0.17510,000 characters

Rate conditions

  1. Approximate published pay-as-you-go rate. Enterprise pricing is custom. This is character billing, not text-token billing.
Pricing source (opens in a new tab)

Data & enterprise terms

Content is retained unless an eligible Enterprise data policy is enabled. Options: no content storage, a 24-hour to 90-day window, or x-expire-content: true for seven-day expiry. On other plans the header is ignored. Billing/usage metadata remains; settings are not retroactive. TTS saved history is excluded from window expiry. Speech-to-speech and cloning are not expressly covered.

Read the service policy (opens in a new tab)

Smallest AI licensed self-host

Provider source (opens in a new tab)
API model ID
Not collected

No pinned inference selector verified for this access route.

Access
Sales-led licensed Docker deployment. Confirm the supplied model artifact and version before assuming parity with the hosted model.
Regions
Customer-managed infrastructure
Deployment mode
Docker; TTS Kubernetes support is not yet documented as availableSource for Deployment mode (opens in a new tab)
Billing basis
Enterprise agreement; no public numeric on-premise inference tariffSource for Billing basis (opens in a new tab)

Endpoint pricing has not been collected. Check the provider for applicable rates.

Data & enterprise terms

Inference runs on customer infrastructure. A license proxy needs outbound access for license validation and usage reporting; do not assume fully air-gapped operation.

Read the service policy (opens in a new tab)

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.

Smallest AI API

lightning_v3.1
Not verified

No discounted prompt-cache contract verified for this model/access route. Audio continuity and model-weight storage are separate features.

How it works
Not collected

No discounted prompt-cache contract verified for this model/access route. Audio continuity and model-weight storage are separate features.

Minimum reusable input
Not collected

No discounted prompt-cache contract verified for this model/access route. Audio continuity and model-weight storage are separate features.

Lifetime & refresh
Not collected

No discounted prompt-cache contract verified for this model/access route. Audio continuity and model-weight storage are separate features.

How to enable it
Not collected

No discounted prompt-cache contract verified for this model/access route. Audio continuity and model-weight storage are separate features.

Usage fields
Not collected

No discounted prompt-cache contract verified for this model/access route. Audio continuity and model-weight storage are separate features.

Billing rules
Not collected

No discounted prompt-cache contract verified for this model/access route. Audio continuity and model-weight storage are separate features.

Limits & data boundaries
Not collected

No discounted prompt-cache contract verified for this model/access route. Audio continuity and model-weight storage are separate features.

No verified cache rate for this endpoint. An empty rate is not a zero charge.

Smallest AI API

lightning_v3.1
Not verified

No discounted prompt-cache contract verified for this model/access route. Audio continuity and model-weight storage are separate features.

How it works
Not collected

No discounted prompt-cache contract verified for this model/access route. Audio continuity and model-weight storage are separate features.

Minimum reusable input
Not collected

No discounted prompt-cache contract verified for this model/access route. Audio continuity and model-weight storage are separate features.

Lifetime & refresh
Not collected

No discounted prompt-cache contract verified for this model/access route. Audio continuity and model-weight storage are separate features.

How to enable it
Not collected

No discounted prompt-cache contract verified for this model/access route. Audio continuity and model-weight storage are separate features.

Usage fields
Not collected

No discounted prompt-cache contract verified for this model/access route. Audio continuity and model-weight storage are separate features.

Billing rules
Not collected

No discounted prompt-cache contract verified for this model/access route. Audio continuity and model-weight storage are separate features.

Limits & data boundaries
Not collected

No discounted prompt-cache contract verified for this model/access route. Audio continuity and model-weight storage are separate features.

No verified cache rate for this endpoint. An empty rate is not a zero charge.

Smallest AI API

lightning_v3.1
Not verified

No discounted prompt-cache contract verified for this model/access route. Audio continuity and model-weight storage are separate features.

How it works
Not collected

No discounted prompt-cache contract verified for this model/access route. Audio continuity and model-weight storage are separate features.

Minimum reusable input
Not collected

No discounted prompt-cache contract verified for this model/access route. Audio continuity and model-weight storage are separate features.

Lifetime & refresh
Not collected

No discounted prompt-cache contract verified for this model/access route. Audio continuity and model-weight storage are separate features.

How to enable it
Not collected

No discounted prompt-cache contract verified for this model/access route. Audio continuity and model-weight storage are separate features.

Usage fields
Not collected

No discounted prompt-cache contract verified for this model/access route. Audio continuity and model-weight storage are separate features.

Billing rules
Not collected

No discounted prompt-cache contract verified for this model/access route. Audio continuity and model-weight storage are separate features.

Limits & data boundaries
Not collected

No discounted prompt-cache contract verified for this model/access route. Audio continuity and model-weight storage are separate features.

No verified cache rate for this endpoint. An empty rate is not a zero charge.

Smallest AI licensed self-host

Not verified

No discounted prompt-cache contract verified for this model/access route. Audio continuity and model-weight storage are separate features.

How it works
Not collected

No discounted prompt-cache contract verified for this model/access route. Audio continuity and model-weight storage are separate features.

Minimum reusable input
Not collected

No discounted prompt-cache contract verified for this model/access route. Audio continuity and model-weight storage are separate features.

Lifetime & refresh
Not collected

No discounted prompt-cache contract verified for this model/access route. Audio continuity and model-weight storage are separate features.

How to enable it
Not collected

No discounted prompt-cache contract verified for this model/access route. Audio continuity and model-weight storage are separate features.

Usage fields
Not collected

No discounted prompt-cache contract verified for this model/access route. Audio continuity and model-weight storage are separate features.

Billing rules
Not collected

No discounted prompt-cache contract verified for this model/access route. Audio continuity and model-weight storage are separate features.

Limits & data boundaries
Not collected

No discounted prompt-cache contract verified for this model/access route. Audio continuity and model-weight storage are separate features.

No verified cache rate for this endpoint. An empty rate is not a zero charge.

Voice choices

234 voices

Choose a voice within Lightning V3.1. These are API voice identifiers, not additional model releases. Source links document each voice.

Published voice catalog
Voice / API IDLanguage & accentDelivery
Yuvikayuvikaenglish, hindi, marathi, gujarati, punjabi, bengali, odia, tamil, telugu, kannada, malayalam
indian		Provider voice label: female. Recommended languages: hindi.Voice source (opens in a new tab)
Siyasiyaenglish, hindi, marathi, gujarati, punjabi, bengali, odia, tamil, telugu, kannada, malayalam
indian		Provider voice label: female. Recommended languages: hindi.Voice source (opens in a new tab)
Prachiprachihindi, english, bengali, punjabi, marathi, gujarati, odia
gujarati		Provider voice label: female. Recommended languages: gujarati.Voice source (opens in a new tab)
Dhruvdhruvenglish, hindi, marathi, gujarati, punjabi, bengali, odia, tamil, telugu, kannada, malayalam
indian		Provider voice label: male. Recommended languages: hindi.Voice source (opens in a new tab)
Nikitanikitaenglish, hindi, marathi, gujarati, punjabi, bengali, odia, tamil, telugu, kannada, malayalam
indian		Provider voice label: female. Recommended languages: hindi.Voice source (opens in a new tab)
Parthparthenglish, hindi, marathi, gujarati, punjabi, bengali, odia, tamil, telugu, kannada, malayalam
indian		Provider voice label: male. Recommended languages: hindi.Voice source (opens in a new tab)
Neelneelenglish, hindi, marathi, gujarati, punjabi, bengali, odia, tamil, telugu, kannada, malayalam
indian		Provider voice label: male. Recommended languages: hindi.Voice source (opens in a new tab)
Devanshdevanshenglish, hindi, marathi, gujarati, punjabi, bengali, odia, tamil, telugu, kannada, malayalam
indian		Provider voice label: male. Recommended languages: hindi.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

Smallest retains rights in its AI models and prohibits model extraction. Its model cards name Smallest as developer; Electron, Hydra and Lightning v3.1 explicitly use proprietary licenses. Commercial self-hosting requires licensed artifacts and a license proxy, not an open-weight grant.

Classification evidence (opens in a new tab)

Sources reviewed