Looking for standalone voice models?
View STT/TTS pricing →↓ Voice agents? Continue below ↓
Voice Agent Pricing
Consumption-based, use case by use case.
Default
Pay as you go
per component / agent minute
Prepaid credit — burns down with usage
Credit rolls forward, never expires
Auto-recharge when your balance runs low
Community support (Discord + GitHub)
Best-effort concurrency, sensible defaults
For those running > 1M call minutes / month
Enterprise
/ month prepaid minimum
Prepaid credit at the same US$ 0.0033 rate
Overage trued up end of month via card on file
Guaranteed concurrency, dedicated capacity
Forward Deployed Engineer + dedicated Slack
P0 / P1 / P2 SLAs · uptime SLA
Custom regions & deployment topology
You're in control
You're in control. Use case by use case.
Tap a layer to switch it between SLNG and your own keys
Execution Layer only
Optimise your existing agent
Keep your orchestrator, STT, LLM, and TTS. SLNG adds the execution layer — cache, routing, governance, marking.
- Execution Layer+ US$ 0.0033
US$ 0.0033/ agent minute
Every component
Pay only for what you use.
Each use case activates only the components it needs. US$ 0.0033 per component, per agent minute.
Routes, caches, governs, auto-optimises, marks and reports. Always active in region.
- ·15 sovereign regions
- ·Data residency controls
- ·Regional PII aware LLM cache, per use case
- ·Regional PII aware TTS cache, per use case
- ·Automatic Call Context Optimization
- ·Automatic Latency Optimisation
- ·PII redaction inline
- ·Prerender TTS + reuse
- ·Per-call boundary decision log
- ·Full call transcripts
- ·Agent Sandbox (10 seconds Active CPU incl.)
- ·SOC 2 · ISO 27001 · HIPAA · GDPR
Any model. Swap freely.
- ·Deepgram, AssemblyAI, Speechmatics, Gladia, Soniox, Whisper, more
- ·Select Provider Per Language
- ·Regional routing per policy
- ·Unified transcript format
- ·Noise Cancelation
Any model. Route by policy. Cache regionally.
- ·GPT-4o, Claude, Gemini, Llama, more
- ·Regional semantic cache (PII Aware)
- ·Policy-based routing
- ·Unified response format
Any model from our Catalogue. Gets faster with every call.
- ·Fish.Audio, Cartesia, Savram, Deepgram, Rime, Soniox plus more
- ·Mid-call voice swap
- ·Regional routing per policy
Framework runtime and carrier connectivity in-region.
- ·LiveKit or Pipecat framework runtime
- ·SIP carrier connectivity, 15 regions
- ·Transfer, hold, DTMF
- ·PSTN + DID management
- ·Worker control channel (autoscaling)
How is SLNG priced?
US$ 0.0033 per agent minute for the execution layer. Same price as every other region. STT and TTS add-ons are each US$ 0.0033 per agent minute with model costs included. No contracts, no minimums, no regional surcharges.
What’s included in every SLNG plan?
Every plan includes access to the same Voice AI Cloud, with all supported models, regions, runtimes, and APIs. SLNG does not gate core capability by plan.
How does usage-based billing work?
Usage is billed independently of your plan. You pay for actual voice usage, while plans and add-ons define concurrency, guarantees, and operational certainty.
How does concurrency pricing work?
Concurrency defines how many calls can run at the same time. Additional capacity can be reserved using add-ons, while usage billing stays the same.
Is SLNG suitable for regulated environments?
Yes. SLNG enforces region and country rules at runtime. Enterprise plans add residency, retention, auditability, and SLA-backed guarantees for regulated workloads.
