Skip to content
Model infoModelsSnapshot: 15 Oct 2026

PlayAI / PlayHT / PlayHT speech

Play 3.0 Mini

Multilingual speech engine with a larger text-input allowance and native 48 kHz generation.

historicalClosed weightsText-to-speech
Official documentation (opens in a new tab)Reviewed 13 Oct 2026
10 of 23 specification fields collected
Text input limit20,000 characters per HTTP streaming requestSee specification notes below
Languages37 listedFull language list in specifications
Input billingNot collectedPlayAI / PlayHT
Output billingNot collectedNo inferred price

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
Streaming
SupportedSource for Streaming (opens in a new tab)

Historical HTTP streaming support; not confirmation of current endpoint availability.

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 37 entries
  • afrikaans
  • albanian
  • amharic
  • arabic
  • bengali
  • bulgarian
  • catalan
  • croatian
  • czech
  • danish
  • dutch
  • english
  • french
  • galician
  • german
  • greek
  • hebrew
  • hindi
  • hungarian
  • indonesian
  • italian
  • japanese
  • korean
  • malay
  • mandarin
  • polish
  • portuguese
  • russian
  • serbian
  • spanish
  • swedish
  • tagalog
  • thai
  • turkish
  • ukrainian
  • urdu
  • xhosa
Source for Languages (opens in a new tab)

The schema enumerates 37 language values. The model overview says 36 for Mini; documentation differs. Voice compatibility still applies.

Context caching
Not collected
Maximum prompt characters
20,000 characters per HTTP streaming requestSource for Maximum prompt characters (opens in a new tab)
Output formats
MP3, μ-law, WAV, OGG, FLACSource for Output formats (opens in a new tab)
Published price
Not collected

No current model-specific price verified. Archived character quotas or third-party estimates are not presented as current rates.

Audio sample rates
48 kHz native; HTTP output sample-rate control 8–48 kHzSource for Audio sample rates (opens in a new tab)
Seed control
Non-negative seed; omit for random seedSource for Seed control (opens in a new tab)
Generation controls
Speed 0.1–5; temperature 0–2; output-quality selectionSource for Generation controls (opens in a new tab)
Audio controls
Emotion presets; voice guidance 1–6; style guidance 1–30Source for Audio controls (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.

API model ID
Play3.0-mini
Access
Historical documented access path. Native service availability has not been verified; surviving documentation is not proof that the route still accepts requests.
Regions
Not established for this historical access path.
API endpoint URL
https://api.play.ht/api/v2/tts/streamSource for API endpoint URL (opens in a new tab)
Streaming
SupportedSource for Streaming (opens in a new tab)

Historical HTTP streaming support; not confirmation of current endpoint availability.

Maximum prompt characters
20,000 characters per HTTP streaming requestSource for Maximum prompt characters (opens in a new tab)
Output formats
MP3, μ-law, WAV, OGG, FLACSource for Output formats (opens in a new tab)

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

Data & enterprise terms

Current retention, training-use and data-residency terms were not verified. Do not infer guarantees from historical API examples.

API model ID
Play3.0-mini
Access
Historical documented access path. Native service availability has not been verified; surviving documentation is not proof that the route still accepts requests.
Regions
Not established for this historical access path.
API endpoint URL
https://api.play.ht/api/v4/websocket-authSource for API endpoint URL (opens in a new tab)

Authentication route returns an expiring WebSocket URL for the selected engine.

Request mode
JSON start frame, binary audio chunks, JSON end frameSource for Request mode (opens in a new tab)
Generation controls
WebSocket examples: speed 0.5–2.0; temperature 0–1Source for Generation controls (opens in a new tab)

Different ranges from HTTP; do not silently substitute HTTP bounds.

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

Data & enterprise terms

Current retention, training-use and data-residency terms were not verified. Do not infer guarantees from historical API examples.

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.

PlayAI / PlayHT

Play3.0-mini
Not verified

No model-specific prompt-cache contract collected. Audio buffering and voice conditioning are not evidence of prompt caching.

How it works
Not collected

No model-specific prompt-cache contract collected. Audio buffering and voice conditioning are not evidence of prompt caching.

Minimum reusable input
Not collected

No model-specific prompt-cache contract collected. Audio buffering and voice conditioning are not evidence of prompt caching.

Lifetime & refresh
Not collected

No model-specific prompt-cache contract collected. Audio buffering and voice conditioning are not evidence of prompt caching.

How to enable it
Not collected

No model-specific prompt-cache contract collected. Audio buffering and voice conditioning are not evidence of prompt caching.

Usage fields
Not collected

No model-specific prompt-cache contract collected. Audio buffering and voice conditioning are not evidence of prompt caching.

Billing rules
Not collected

No model-specific prompt-cache contract collected. Audio buffering and voice conditioning are not evidence of prompt caching.

Limits & data boundaries
Not collected

No model-specific prompt-cache contract collected. Audio buffering and voice conditioning are not evidence of prompt caching.

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

PlayAI / PlayHT

Play3.0-mini
Not verified

No model-specific prompt-cache contract collected. Audio buffering and voice conditioning are not evidence of prompt caching.

How it works
Not collected

No model-specific prompt-cache contract collected. Audio buffering and voice conditioning are not evidence of prompt caching.

Minimum reusable input
Not collected

No model-specific prompt-cache contract collected. Audio buffering and voice conditioning are not evidence of prompt caching.

Lifetime & refresh
Not collected

No model-specific prompt-cache contract collected. Audio buffering and voice conditioning are not evidence of prompt caching.

How to enable it
Not collected

No model-specific prompt-cache contract collected. Audio buffering and voice conditioning are not evidence of prompt caching.

Usage fields
Not collected

No model-specific prompt-cache contract collected. Audio buffering and voice conditioning are not evidence of prompt caching.

Billing rules
Not collected

No model-specific prompt-cache contract collected. Audio buffering and voice conditioning are not evidence of prompt caching.

Limits & data boundaries
Not collected

No model-specific prompt-cache contract collected. Audio buffering and voice conditioning are not evidence of prompt caching.

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

Named provider-operated speech engine with no public weight distribution identified in the reviewed official inventory. PlayAI-Groq documents a commercial model license for Dialog. Classification of legacy engines is based on provider-controlled access, not a claim that all PlayAI work is closed; public PlayDiffusion checkpoints are explicitly excluded.

Classification evidence (opens in a new tab)

Sources reviewed