Cartesia · Rate Limits

Cartesia Ai Rate Limits

Cartesia enforces concurrency-based limits rather than classic requests-per-minute quotas. Each plan caps the number of simultaneous TTS requests, simultaneous STT requests, and active agent slots; a single TTS WebSocket connection can multiplex "dozens" of concurrent generation contexts within its plan's TTS concurrency ceiling. Usage above the plan's monthly credit allowance is billed or blocked depending on account configuration.

Cartesia Ai Rate Limits is the machine-readable rate-limit profile for Cartesia on the APIs.io network, conforming to the API Commons Rate Limits specification.

It captures 15 rate-limit definitions, measuring connections, agents, contexts, frames, and seconds.

The profile also includes 3 backoff/retry policies defined and response codes documented for throttled.

Tagged areas include AI, Voice AI, Text to Speech, Speech to Text, and WebSocket.

15 Limits Throttle: 429
AIVoice AIText to SpeechSpeech to TextWebSocketRate LimitingQuotasConcurrency

Limits

Concurrent TTS Requests (Free) account
connections
2
Applies to /tts/bytes, /tts/sse, and /tts/websocket combined.
Concurrent TTS Requests (Pro) account
connections
3
Applies to /tts/bytes, /tts/sse, and /tts/websocket combined.
Concurrent TTS Requests (Startup) account
connections
5
Applies to /tts/bytes, /tts/sse, and /tts/websocket combined.
Concurrent TTS Requests (Scale) account
connections
15
Applies to /tts/bytes, /tts/sse, and /tts/websocket combined.
Concurrent STT Requests (Free) account
connections
8
Applies to /stt, /stt/websocket, and /stt/turns/websocket combined.
Concurrent STT Requests (Pro) account
connections
12
Applies to /stt, /stt/websocket, and /stt/turns/websocket combined.
Concurrent STT Requests (Startup) account
connections
20
Applies to /stt, /stt/websocket, and /stt/turns/websocket combined.
Concurrent STT Requests (Scale) account
connections
60
Applies to /stt, /stt/websocket, and /stt/turns/websocket combined.
Agent Slots (Free) account
agents
1
Number of configured voice agents allowed simultaneously.
Agent Slots (Pro) account
agents
3
Number of configured voice agents allowed simultaneously.
Agent Slots (Startup) account
agents
5
Number of configured voice agents allowed simultaneously.
Agent Slots (Scale) account
agents
10
Number of configured voice agents allowed simultaneously.
TTS WebSocket Context Multiplexing connection
contexts
dozens per connection (published guidance, not a hard number)
A single wss://api.cartesia.ai/tts/websocket connection can run many concurrent context_id generations; still bounded by the account's overall concurrent TTS request limit.
STT WebSocket Audio Cadence connection
frames
100ms recommended chunk size
Binary audio frames should be sent in roughly 100ms chunks matching the negotiated encoding and sample_rate.
Access Token Lifetime token
seconds
3600
Short-lived scoped tokens minted via POST /access-token expire after at most 3600 seconds.

Policies

Concurrency Over Rate
Cartesia governs load primarily through simultaneous connection/request ceilings per plan rather than a rolling per-minute request counter.
Enterprise Custom Concurrency
Enterprise agreements can raise TTS/STT concurrency and agent slot limits beyond the Scale tier.
Backoff Strategy
Clients should implement exponential backoff with jitter and honor Retry-After on 429 responses, and should reuse a single WebSocket connection across many contexts rather than opening one per generation.

Sources