Languages31 listedFull language list in specifications
Input billingSee ratesLMNT API (retired)
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
No model-specific specification verified. Unknown does not mean unsupported.
Maximum input tokens
Not collected
No model-specific specification verified. Unknown does not mean unsupported.
Maximum output
Not collected
No model-specific specification verified. Unknown does not mean unsupported.
Parameters
Not collected
No model-specific specification verified. Unknown does not mean unsupported.
Release date
Not collected
Family experimental release: December 2024; production default: May 2025. Exact Blizzard 1.4 release date is not established by those family announcements or the archive capture date.
Knowledge cutoff
Not collected
No model-specific specification verified. Unknown does not mean unsupported.
Tool calling
Not collected
No model-specific specification verified. Unknown does not mean unsupported.
Structured output
Not collected
No model-specific specification verified. Unknown does not mean unsupported.
Binary HTTP speech and WebSocket speech sessions; AAC and WAV wait for full generation.
Batch processing
Not collected
No model-specific specification verified. Unknown does not mean unsupported.
Fine-tuning
Not collected
No model-specific specification verified. Unknown does not mean unsupported.
Reasoning
Not collected
No model-specific specification verified. Unknown does not mean unsupported.
Context caching
Not collected
No model-specific specification verified. Unknown does not mean unsupported.
Base model
Not collected
No model-specific specification verified. Unknown does not mean unsupported.
Endpoint availability
LMNT has shut down. Historical contracts below do not establish a callable service. Exact shutdown date and any continuing third-party deployment are unverified.Source for Endpoint availability (opens in a new tab)
Creation privacy
Archived June 2023 policy allows user-content use for machine-learning training and human review. Personal information is generally retained until edited/deleted or while the person remains a user; no fixed inference-retention period or zero-retention promise established. US and other international transfers are described. Post-shutdown deletion and handling are unverified.Source for Creation privacy (opens in a new tab)
Model terms
Archived site terms describe personal, noncommercial site access; historical paid API plans advertise commercial licenses. These have different scopes, so neither a universal commercial prohibition nor unrestricted commercial rights is inferred. Separate service restrictions protect model weights. Current contractual status after shutdown is unknown.Source for Model terms (opens in a new tab)
blizzard was the API selector for the Blizzard family; the archived overview identifies version 1.4. No immutable 1.4 selector or earlier numbered checkpoints verified.Source for Model selection (opens in a new tab)
31 languages plus auto detection. Native scripts and explicit language selection are recommended. Code switching is documented; language-based accent control does not translate input text.Source for Language coverage (opens in a new tab)
Reference voice and language guide accent; temperature controls expressive variation and top_p stability. Current SDK gives examples but no universal numeric bounds for these floats.Source for Voice controls (opens in a new tab)
HTTP X-API-Key and lmnt-version compatibility header; sessions send credentials in their initialization JSON. An API version is not a model version.Source for Authentication (opens in a new tab)
Response format
Detailed speech returns base64 audio and optional start/duration/text elements, including words, spaces and punctuation. Protocol JSON is not caller-defined structured generation.Source for Response format (opens in a new tab)
Documentation discrepancy
Earlier v2.0.0 SDK includes seed and raw output with 19 explicit languages plus auto. Reviewed later schema has 31 languages, pcm_s16le/pcm_f32le and debug but no seed. µ-law prose says mulaw while the format enum says ulaw. Do not merge these into one timeless contract.Source for Documentation discrepancy (opens in a new tab)
Published price
Historical API subscription bundles with character overages. August 2026 archive: Free USD 0 with 15K characters; Indie USD 10/month with 200K; Pro USD 49/month with 1.25M; Premium USD 199/month with 5.7M. Paid overages vary by plan. Not currently purchasable or token-based.Source for Published price (opens in a new tab)
Request quotas
Historical paid plans advertise no concurrency or rate limits. No separately verified per-model SLA, account-specific restrictions or current availability.Source for Request quotas (opens in a new tab)
Billing basis
Characters per billing period. SDK usage.characters means remaining characters, not consumed characters; plan exposes character_limit and commercial_use_allowed. Billing rounding and cancellation charges unverified.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.
Rolling family selector; not an immutable version pin.
Access
LMNT has shut down. Historical contracts below do not establish a callable service. Exact shutdown date and any continuing third-party deployment are unverified. Historically POST https://api.lmnt.com/v1/ai/speech/bytes with text and voice. Read binary audio; streamable formats emit chunks.
Regions
Historical inference residency unverified. Company location is not processing-region evidence.
HTTP speech request limit including spaces. Not a token count or a whole-session lifetime limit.
Prices retain their published currencies and billing units below.
Published billing units · original currency · observed 16 Oct 2026
Category
Usage
Rate
Unit
Historical API subscriptions
Free: 15K included characters1
$0.00
free plan allowance
Historical API subscriptions
Indie: 200K included characters2
$10.00
subscription month
Historical character overages
Indie: characters after included allowance3
$0.05
1,000 characters
Historical API subscriptions
Pro: 1.25M included characters2
$49.00
subscription month
Historical character overages
Pro: characters after included allowance4
$0.045
1,000 characters
Historical API subscriptions
Premium: 5.7M included characters2
$199.00
subscription month
Historical character overages
Premium: characters after included allowance5
$0.035
1,000 characters
Rate conditions
Archived August 3, 2026; provider has since shut down. Free allowance reset cadence not established by this page. Unlimited voice clones are an included platform benefit. Same plan across alternative speech routes, not additive charges. No token/question conversion.
Archived August 3, 2026; provider has since shut down. Monthly plan; paid plans advertise commercial license and no concurrency or rate limits. Unlimited voice clones are an included platform benefit. Same plan across alternative speech routes, not additive charges. No token/question conversion.
Requires historical Indie subscription and applies only after its included allowance. Not standalone pay-as-you-go; not currently purchasable. Same tariff across alternative speech routes, not additive charges. No text-token or question estimate.
Requires historical Pro subscription and applies only after its included allowance. Not standalone pay-as-you-go; not currently purchasable. Same tariff across alternative speech routes, not additive charges. No text-token or question estimate.
Requires historical Premium subscription and applies only after its included allowance. Not standalone pay-as-you-go; not currently purchasable. Same tariff across alternative speech routes, not additive charges. No text-token or question estimate.
Archived June 2023 policy allows user-content use for machine-learning training and human review. Personal information is generally retained until edited/deleted or while the person remains a user; no fixed inference-retention period or zero-retention promise established. US and other international transfers are described. Post-shutdown deletion and handling are unverified. Archived site terms describe personal, noncommercial site access; historical paid API plans advertise commercial licenses. These have different scopes, so neither a universal commercial prohibition nor unrestricted commercial rights is inferred. Separate service restrictions protect model weights. Current contractual status after shutdown is unknown.
Rolling family selector; not an immutable version pin.
Access
LMNT has shut down. Historical contracts below do not establish a callable service. Exact shutdown date and any continuing third-party deployment are unverified. Historically POST https://api.lmnt.com/v1/ai/speech for base64 audio and optional timestamps. Whole response waits for completed synthesis.
Regions
Historical inference residency unverified. Company location is not processing-region evidence.
HTTP lazy-response wrappers do not make this a realtime audio operation.
Response format
Detailed speech returns base64 audio and optional start/duration/text elements, including words, spaces and punctuation. Protocol JSON is not caller-defined structured generation.Source for Response format (opens in a new tab)
Prices retain their published currencies and billing units below.
Published billing units · original currency · observed 16 Oct 2026
Category
Usage
Rate
Unit
Historical API subscriptions
Free: 15K included characters1
$0.00
free plan allowance
Historical API subscriptions
Indie: 200K included characters2
$10.00
subscription month
Historical character overages
Indie: characters after included allowance3
$0.05
1,000 characters
Historical API subscriptions
Pro: 1.25M included characters2
$49.00
subscription month
Historical character overages
Pro: characters after included allowance4
$0.045
1,000 characters
Historical API subscriptions
Premium: 5.7M included characters2
$199.00
subscription month
Historical character overages
Premium: characters after included allowance5
$0.035
1,000 characters
Rate conditions
Archived August 3, 2026; provider has since shut down. Free allowance reset cadence not established by this page. Unlimited voice clones are an included platform benefit. Same plan across alternative speech routes, not additive charges. No token/question conversion.
Archived August 3, 2026; provider has since shut down. Monthly plan; paid plans advertise commercial license and no concurrency or rate limits. Unlimited voice clones are an included platform benefit. Same plan across alternative speech routes, not additive charges. No token/question conversion.
Requires historical Indie subscription and applies only after its included allowance. Not standalone pay-as-you-go; not currently purchasable. Same tariff across alternative speech routes, not additive charges. No text-token or question estimate.
Requires historical Pro subscription and applies only after its included allowance. Not standalone pay-as-you-go; not currently purchasable. Same tariff across alternative speech routes, not additive charges. No text-token or question estimate.
Requires historical Premium subscription and applies only after its included allowance. Not standalone pay-as-you-go; not currently purchasable. Same tariff across alternative speech routes, not additive charges. No text-token or question estimate.
Archived June 2023 policy allows user-content use for machine-learning training and human review. Personal information is generally retained until edited/deleted or while the person remains a user; no fixed inference-retention period or zero-retention promise established. US and other international transfers are described. Post-shutdown deletion and handling are unverified. Archived site terms describe personal, noncommercial site access; historical paid API plans advertise commercial licenses. These have different scopes, so neither a universal commercial prohibition nor unrestricted commercial rights is inferred. Separate service restrictions protect model weights. Current contractual status after shutdown is unknown.
No immutable model selector verified. Voice IDs identify profiles, not base checkpoints.
Access
LMNT has shut down. Historical contracts below do not establish a callable service. Exact shutdown date and any continuing third-party deployment are unverified. Historically wss://api.lmnt.com/v1/ai/speech/stream. Init includes X-API-Key, voice, compatibility version and optional format/language/sample rate/timestamps. No model selector in reviewed init.
Regions
Historical inference residency unverified. Company location is not processing-region evidence.
Binary audio plus ready, timestamps, flush_complete, reset_complete or error JSON. Flush generates buffered text and retains connection; reset drops buffered text; finish requests closure after dispatch. Flush/reset use nonce acknowledgements. Errors carry request_id. Cancellation billing and maximum session lifetime unknown.Source for Response format (opens in a new tab)
Output limits and timing
Timestamp start times are cumulative and reset to zero after flush or reset. Per-session text ceiling is unverified; the 5,000-character HTTP limit is not assigned to entire sessions.Source for Output limits and timing (opens in a new tab)
Prices retain their published currencies and billing units below.
Published billing units · original currency · observed 16 Oct 2026
Category
Usage
Rate
Unit
Historical API subscriptions
Free: 15K included characters1
$0.00
free plan allowance
Historical API subscriptions
Indie: 200K included characters2
$10.00
subscription month
Historical character overages
Indie: characters after included allowance3
$0.05
1,000 characters
Historical API subscriptions
Pro: 1.25M included characters2
$49.00
subscription month
Historical character overages
Pro: characters after included allowance4
$0.045
1,000 characters
Historical API subscriptions
Premium: 5.7M included characters2
$199.00
subscription month
Historical character overages
Premium: characters after included allowance5
$0.035
1,000 characters
Rate conditions
Archived August 3, 2026; provider has since shut down. Free allowance reset cadence not established by this page. Unlimited voice clones are an included platform benefit. Same plan across alternative speech routes, not additive charges. No token/question conversion.
Archived August 3, 2026; provider has since shut down. Monthly plan; paid plans advertise commercial license and no concurrency or rate limits. Unlimited voice clones are an included platform benefit. Same plan across alternative speech routes, not additive charges. No token/question conversion.
Requires historical Indie subscription and applies only after its included allowance. Not standalone pay-as-you-go; not currently purchasable. Same tariff across alternative speech routes, not additive charges. No text-token or question estimate.
Requires historical Pro subscription and applies only after its included allowance. Not standalone pay-as-you-go; not currently purchasable. Same tariff across alternative speech routes, not additive charges. No text-token or question estimate.
Requires historical Premium subscription and applies only after its included allowance. Not standalone pay-as-you-go; not currently purchasable. Same tariff across alternative speech routes, not additive charges. No text-token or question estimate.
Archived June 2023 policy allows user-content use for machine-learning training and human review. Personal information is generally retained until edited/deleted or while the person remains a user; no fixed inference-retention period or zero-retention promise established. US and other international transfers are described. Post-shutdown deletion and handling are unverified. Archived site terms describe personal, noncommercial site access; historical paid API plans advertise commercial licenses. These have different scopes, so neither a universal commercial prohibition nor unrestricted commercial rights is inferred. Separate service restrictions protect model weights. Current contractual status after shutdown is unknown.
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.
LMNT API (retired)
blizzard
Not verified
No inference-cache contract verified. Persistent speech sessions, stored voice prompts, saved clips, SDK cached properties and random seeds are not discounted input caching.
How it works
Not collected
No inference-cache contract verified. Persistent speech sessions, stored voice prompts, saved clips, SDK cached properties and random seeds are not discounted input caching.
Minimum reusable input
Not collected
No inference-cache contract verified. Persistent speech sessions, stored voice prompts, saved clips, SDK cached properties and random seeds are not discounted input caching.
Lifetime & refresh
Not collected
No inference-cache contract verified. Persistent speech sessions, stored voice prompts, saved clips, SDK cached properties and random seeds are not discounted input caching.
How to enable it
Not collected
No inference-cache contract verified. Persistent speech sessions, stored voice prompts, saved clips, SDK cached properties and random seeds are not discounted input caching.
Usage fields
Not collected
No inference-cache contract verified. Persistent speech sessions, stored voice prompts, saved clips, SDK cached properties and random seeds are not discounted input caching.
Billing rules
Not collected
No inference-cache contract verified. Persistent speech sessions, stored voice prompts, saved clips, SDK cached properties and random seeds are not discounted input caching.
Limits & data boundaries
Not collected
No inference-cache contract verified. Persistent speech sessions, stored voice prompts, saved clips, SDK cached properties and random seeds are not discounted input caching.
No verified cache rate for this endpoint. An empty rate is not a zero charge.
Caching reviewed 16 Oct 2026
LMNT API (retired)
blizzard
Not verified
No inference-cache contract verified. Persistent speech sessions, stored voice prompts, saved clips, SDK cached properties and random seeds are not discounted input caching.
How it works
Not collected
No inference-cache contract verified. Persistent speech sessions, stored voice prompts, saved clips, SDK cached properties and random seeds are not discounted input caching.
Minimum reusable input
Not collected
No inference-cache contract verified. Persistent speech sessions, stored voice prompts, saved clips, SDK cached properties and random seeds are not discounted input caching.
Lifetime & refresh
Not collected
No inference-cache contract verified. Persistent speech sessions, stored voice prompts, saved clips, SDK cached properties and random seeds are not discounted input caching.
How to enable it
Not collected
No inference-cache contract verified. Persistent speech sessions, stored voice prompts, saved clips, SDK cached properties and random seeds are not discounted input caching.
Usage fields
Not collected
No inference-cache contract verified. Persistent speech sessions, stored voice prompts, saved clips, SDK cached properties and random seeds are not discounted input caching.
Billing rules
Not collected
No inference-cache contract verified. Persistent speech sessions, stored voice prompts, saved clips, SDK cached properties and random seeds are not discounted input caching.
Limits & data boundaries
Not collected
No inference-cache contract verified. Persistent speech sessions, stored voice prompts, saved clips, SDK cached properties and random seeds are not discounted input caching.
No verified cache rate for this endpoint. An empty rate is not a zero charge.
Caching reviewed 16 Oct 2026
LMNT API (retired)
Not verified
No inference-cache contract verified. Persistent speech sessions, stored voice prompts, saved clips, SDK cached properties and random seeds are not discounted input caching.
How it works
Not collected
No inference-cache contract verified. Persistent speech sessions, stored voice prompts, saved clips, SDK cached properties and random seeds are not discounted input caching.
Minimum reusable input
Not collected
No inference-cache contract verified. Persistent speech sessions, stored voice prompts, saved clips, SDK cached properties and random seeds are not discounted input caching.
Lifetime & refresh
Not collected
No inference-cache contract verified. Persistent speech sessions, stored voice prompts, saved clips, SDK cached properties and random seeds are not discounted input caching.
How to enable it
Not collected
No inference-cache contract verified. Persistent speech sessions, stored voice prompts, saved clips, SDK cached properties and random seeds are not discounted input caching.
Usage fields
Not collected
No inference-cache contract verified. Persistent speech sessions, stored voice prompts, saved clips, SDK cached properties and random seeds are not discounted input caching.
Billing rules
Not collected
No inference-cache contract verified. Persistent speech sessions, stored voice prompts, saved clips, SDK cached properties and random seeds are not discounted input caching.
Limits & data boundaries
Not collected
No inference-cache contract verified. Persistent speech sessions, stored voice prompts, saved clips, SDK cached properties and random seeds are not discounted input caching.
No verified cache rate for this endpoint. An empty rate is not a zero charge.
Caching reviewed 16 Oct 2026
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
Official model overview and release notes identify LMNT speech models; service policy explicitly restricts discovery of its models and model weights. Open client code and unrelated public DiffWave/WaveGrad research do not establish published Aurora or Blizzard weights.
