> ## Documentation Index > Fetch the complete documentation index at: https://www.vobiz.ai/docs/llms.txt > Use this file to discover all available pages before exploring further. # Vobiz vs Retell AI — Telephony Infrastructure vs Voice Agent Platform > Retell AI is an AI voice agent orchestration platform. Vobiz is telephony infrastructure. They're complementary — Retell AI runs on top of Vobiz SIP trunking. How they fit together and when to use each. ## Vobiz vs Retell AI: Different Layers of the Stack Vobiz and Retell AI operate at different layers. Retell AI is an AI voice agent orchestration platform — it handles the LLM, TTS, conversation logic, and agent configuration. Vobiz is telephony infrastructure — it handles SIP trunking, phone numbers, inbound/outbound call routing, AMD, and audio streaming. **They are designed to work together, not replace each other.** Retell AI's recommended setup for PSTN calling is to connect a SIP trunk from a provider like Vobiz. If you're looking for a telephony layer for Retell AI, Vobiz is the answer. If you're looking to replace Retell AI's orchestration layer, that's a different comparison. ## What each platform does | Layer | Retell AI | Vobiz | | - | - | - | | **LLM (conversation brain)** | ✓ Managed, any model | — | | **TTS (voice synthesis)** | ✓ Multiple providers | — | | **STT (transcription)** | ✓ Managed | — | | **Agent logic / flows** | ✓ Dashboard and API | — | | **SIP trunking** | Requires external provider | ✓ | | **Phone numbers** | Requires external provider (or limited built-in US/CA) | ✓ 130+ countries | | **India 140-series numbers** | Not available built-in | ✓ | | **AMD (answering machine detection)** | Via SIP trunk | ✓ Per-call | | **Audio streaming (WebSocket)** | ✓ To your SIP trunk | ✓ Bidirectional | | **Call recordings** | — | ✓ | | **XML call control** | — | ✓ | | **Developer API** | ✓ | ✓ | | **Self-serve signup** | ✓ | ✓ | ## How they work together The integration is straightforward. Vobiz provides the SIP trunk and phone numbers; Retell AI provides the agent logic, STT, LLM, and TTS. When a call comes in or goes out, Vobiz routes it to Retell AI via SIP. ``` Incoming call (PSTN) │ ▼ Vobiz SIP trunk │ (SIP INVITE) ▼ Retell AI agent ├── STT: speech → text ├── LLM: generate response └── TTS: text → speech │ (audio back via SIP) ▼ Caller hears the AI agent ``` For outbound calls, Retell AI instructs Vobiz (via REST API or SIP) to place the call. Vobiz dials, runs AMD, and connects the answered call to the Retell AI agent session. See the [Retell AI + Vobiz integration guide](https://www.vobiz.ai/docs/integrations/retellai-api.md) for the full setup. ## When to use Retell AI + Vobiz Use Retell AI as your orchestration layer and Vobiz as your telephony layer when: * You want a managed agent dashboard with pre-built flows, testing tools, and analytics * You need multiple LLM and TTS provider options through one interface * Your team prefers a low-code/no-code way to configure agent behavior * You need PSTN calling in countries beyond the US and Canada (Vobiz covers 130+) * You need India 140-series numbers for TRAI-compliant A2P outbound calling ## When to use Vobiz alone (without Retell AI) Vobiz's XML call control and WebSocket audio streaming let you build voice AI agents without Retell AI as the orchestration layer. Use Vobiz directly when: * You want to control every layer: your own STT, LLM, TTS, and conversation logic * You're integrating with an orchestration layer that isn't Retell AI (Vapi, Pipecat, LiveKit, ElevenLabs) * You want to combine IVR flows (XML) with AI segments in the same call * You need maximum flexibility in audio codec and streaming format ```xml Example: Vobiz XML with embedded AI stream theme={null} Welcome to Acme support. wss://your-ai-agent.com/stream ``` ## Pricing comparison | | Vobiz | Retell AI | | - | - | - | | **Model** | Pay-as-you-go telephony | Per-minute AI agent + telephony | | **Voice rate (India)** | ₹0.45/min | Varies by plan + SIP trunk cost | | **Phone numbers** | Per number per month | Per number per month (limited countries) | | **Free tier** | 3 concurrent calls + free credits on signup | Trial credits | | **Billing currency** | INR (GST invoice) | USD | | **Concurrency** | Scales to unlimited | Plan-limited | When using Retell AI + Vobiz, you pay Retell AI for the agent minutes (LLM + TTS + orchestration) and Vobiz for the telephony minutes and number rental. The two costs are separate invoices. ## India support comparison | | Vobiz | Retell AI | | - | - | - | | **India phone numbers** | ✓ DID, 140-series | ✗ Not available | | **India outbound** | ✓ Direct interconnects | Depends on SIP trunk | | **140-series caller ID** | ✓ | ✗ | | **INR billing** | ✓ | ✗ (USD only) | | **GST invoice** | ✓ | ✗ | | **India data residency** | Contact support | ✗ | For teams building AI voice agents in India, Vobiz handles the TRAI-compliance layer (140-series numbers, DLT registration, NDNC compliance) that Retell AI doesn't cover. ## Developer experience Both platforms are API-first with good documentation. The key difference is scope: * **Retell AI docs** cover agent configuration, LLM prompt setup, testing, and analytics * **Vobiz docs** cover call routing, SIP trunking, number management, webhooks, and audio streaming For the integration between them, see the [Retell AI integration guide](https://www.vobiz.ai/docs/integrations/retellai-api.md) (SIP trunk import and outbound call setup) and the [Retell AI dashboard integration](https://www.vobiz.ai/docs/integrations/retellai-dashboard.md) (UI-based setup). ## Summary | Use case | Recommendation | | - | - | | Need an AI agent platform with dashboard | Retell AI + Vobiz for telephony | | Need India 140-series numbers | Vobiz (required either way) | | Building a custom AI pipeline | Vobiz alone (bring your own STT/LLM/TTS) | | Calling 130+ countries | Vobiz (Retell AI's built-in numbers are US/CA only) | | Switching SIP trunk provider for Retell AI | Switch to Vobiz | | Replacing Retell AI's orchestration | Use a different orchestration layer with Vobiz | This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.