Plivo vs Telnyx (2026)

Flows act in other systems through app action nodes with prebuilt actions for Zendesk, HubSpot and Google Workspace, HTTP request nodes for any API, and call forward nodes.

The wider catalog adds Salesforce, Zoho CRM, Shopify, Stripe, Intercom, Jira, Notion, Airtable, Freshdesk, Slack, Google Calendar, Calendly, ServiceNow, GitHub, Google Sheets, MongoDB, BigQuery and PostgreSQL, along with MCP servers, API based tools and Python code nodes. In a single conversation an agent can look up a booking, check availability, book a slot and text a confirmation.

Apps are authenticated once under Settings and then used across agents, and the team handles that authentication and any field mapping itself before the Vibe Agent can use an app in a draft.

Sourceplivo.com/docs/aiagent/getstarted/overviewread 2026-10-11

Assistants call outside systems with webhook tools, which can run asynchronously with filler messages while the request completes, attach MCP servers for whole tool catalogs, and connect enterprise integrations for Salesforce, Zendesk and Jira. Call control tools transfer the caller, hand off with SIP Refer, send DTMF tones to another phone system and hang up, client side tools run actions in the web widget, and keys for outside providers are kept as write only integration secrets. Tools live in a shared library that several assistants and workflow nodes can reuse.

Sourcedevelopers.telnyx.com/docs/inference/ai-assistants/no-code-voice-assistant.mdread 2026-10-11

Each agent is a flow on a drag and drop canvas. Nodes such as an AI reply, an app action, an HTTP request or a call forward are joined by paths that branch on the customer's input or on system conditions, and variables carry extracted or generated values from step to step. A trigger node starts the flow from a channel or an API call.

The Vibe Agent drafts a flow from a plain English description, first writing a plan that covers the trigger, channel setup, node settings, edge cases and when to close the call, asking clarifying questions and flagging requests it cannot build. By Plivo's own estimate the draft gets a flow about 70 percent of the way, and every generated node is meant to be reviewed by hand. The Playground highlights the active node and the path taken through the flow.

Sourceplivo.com/docs/aiagent/getstarted/overviewread 2026-10-11

Conversation workflows store a directed graph on the assistant. Prompt nodes run the model with their own instructions and optional model and voice overrides, speak nodes play a fixed message, and tool nodes run one shared tool and advance.

Edges move between nodes on a natural language condition the model judges, a deterministic comparison against variables, or a default, and a tool node's HTTP status code can steer the next edge.

An edge can also hand the conversation to another assistant, so a triage assistant can route billing, technical and sales callers to specialists that share the conversation's context.

Sourcedevelopers.telnyx.com/docs/inference/ai-assistants/workflows.mdread 2026-10-11

Flows start when a call, a web chat message, an SMS or a WhatsApp message arrives, or when another system sends an authenticated HTTP request with optional input parameters, with basic auth on by default.

One agent can go live on voice through a purchased number, WhatsApp through a business account with template management, web chat and SMS over 10DLC, and conversations run inbound and outbound.

Each number takes one inbound flow while one flow can answer several numbers, outbound calls start from an Initiate call node or an API request, and the flow branches on whether a call was answered, busy, unanswered or rejected.

Sourceplivo.com/docs/aiagent/aistudio/triggerread 2026-10-11

Calls reach an assistant on an assigned Telnyx number or over SIP, outbound calls start from the portal, the API or the TeXML ai_calls endpoint, and an embeddable widget carries voice and chat on a web page. SMS conversations run through the assistant SMS chat API, with MMS during a call when messaging is on. At the start of each conversation a dynamic variables webhook can load caller details and memory settings, and custom SIP headers or the outbound request can pass variables in.

Sourcedevelopers.telnyx.com/docs/inference/ai-assistants/no-code-voice-assistant.mdread 2026-10-11

A knowledge base combines uploaded files of up to 64MB, including spreadsheets, PDFs, Word documents, text, JSON and XML, whole websites of up to 10,000 URLs with limits on depth and paths, and apps such as Notion, Confluence, Freshdesk, Zendesk and Intercom. Website and app sources refresh automatically as often as weekly, one knowledge base can serve several agents, and the agent runs a semantic search at every response step, passing the most relevant chunks to the model and pulling nothing when nothing matches.

Sourceplivo.com/docs/aiagent/aistudio/agentconfiguration/knowledgebaseread 2026-10-11

Each assistant can use named knowledge bases built from uploaded files or a URL, kept with the account and attached in the assistant's configuration, so one knowledge base can serve several assistants. Retrieval during a call is part of the voice engine's orchestration and included in the per minute price, while the price estimator carries a separate agent knowledge base add on. Telnyx does not say how the content is chunked or indexed.

Sourcetelnyx.com/pricing/conversational-airead 2026-10-11

Variables hold names, emails, IDs and other values the agent extracts while a flow runs. They are set on the AI conversation node's Extract Variables tab and passed to later steps, and each reply draws on the path the conversation has taken so far. Plivo says context carries over when one customer moves from one channel to another, though where that context is kept, for how long, and whether it reaches a separate conversation days later is left open.

Sourceplivo.com/docs/aiagent/getstarted/overviewread 2026-10-11

Memory brings earlier conversations into a new one. The dynamic variables webhook returns a query in the same form as the List Conversations endpoint, typically the caller's phone number, a time window or custom metadata, plus the IDs of insights to include, and the assistant starts with those conversations and insight results in context, across voice and messaging.

The webhook has one second to answer before the call carries on without memory. What comes back is stored conversation history picked by the customer's query. There is no separate memory store, and Telnyx sets out no retention period or way to delete one person's history.

Sourcedevelopers.telnyx.com/docs/inference/ai-assistants/memoryread 2026-10-11

Conversations hand over to human specialists assigned by agent, team or custom rules and routed by skills or queues, with business hours, holidays, agent status, SLAs, retries and workload managed from one dashboard, and calls transfer to a person by number forwarding or authenticated SIP transfer. Transfer to a human is included in the agent price. No step lets a person approve an action or a reply before the agent sends it.

Sourceplivo.com/docs/aiagent/getstarted/overviewread 2026-10-11

Transfer and SIP Refer tools pass a caller to a person or another phone system, with voicemail detection on transfer, and handoff tools move a conversation between assistants. Releases stay under the team's control. A new version is saved with notes, traffic rules send selected callers or a percentage to it, and rollback returns all traffic to the main version at once. No step lets a person approve a tool call or a reply before the assistant acts.

Sourcedevelopers.telnyx.com/docs/inference/ai-assistants/version-testing-traffic-distribution.mdread 2026-10-11

Plivo holds a SOC 2 certification of unstated type, audited each year by an independent firm, is certified compliant with PCI DSS Level 1, has completed a CSA STAR Level 1 self assessment, takes part in the EU and US Data Privacy Framework, and signs a HIPAA business associate agreement on Enterprise.

Each user signs in with unique credentials, a passphrase and a phone verification code sent by SMS or voice call, data is encrypted with TLS 1.2 or higher in transit and 256 bit AES at rest, and backups are encrypted and stored redundantly. Single sign on comes with the Enterprise plan, and Plivo says nothing of roles or permissions inside the studio.

Sourceplivo.com/securityread 2026-10-11

Business accounts sign in to the Mission Control Portal through SAML single sign on with Okta, Azure Active Directory, Google, OneLogin, Auth0 or LastPass. Telnyx holds ISO 27001:2022 and ISO 27701, a SOC 2 Type II report under NDA covering voice, messaging, video and wireless, and HIPAA coverage for voice, messaging, fax and AI inference with a BAA available, plus PCI DSS for its own billing. The SOC 2 scope does not name AI assistants, and Telnyx also claims a SOC 2 Type III, a report type the AICPA does not issue.

Sourcetelnyx.com/ai/compliance.mdread 2026-10-11

Six reporting dashboards cover conversations, customer metrics, agent and AI performance, availability and operations. In the Playground, analysis mode shows each node's execution with the prompt used, the tool response, response time and model metrics, and every session carries a run ID for tracing it in the logs. Outside the Playground, node by node detail on live conversations and a log of who changed an agent go unmentioned.

Sourceplivo.com/docs/aiagent/aistudio/playgroundread 2026-10-11

Every conversation is stored and listable through the conversations API, transcripts show which workflow node produced each assistant message, and the last tool call's HTTP status is kept as a variable. Insight groups run structured and unstructured analysis on each conversation and can post results to a webhook, with Telnyx managed insights alongside custom ones. Version notes record what changed in each release, and Telnyx does not mention an audit trail of who changed an assistant's configuration.

Sourcedevelopers.telnyx.com/docs/inference/ai-assistants/workflows.mdread 2026-10-11

Plivo advertises data residency in the US, the EU and Asia Pacific, while its platform runs on Amazon Web Services compute and storage spread worldwide. Which region an account lands in, and whether the customer chooses it, is not spelled out. The studio runs as a hosted service only, with no private cloud or on premises edition.

Sourceplivo.comread 2026-10-11

Voice AI agents run in the United States, the primary data center, or in the European Union with EU data residency, where data does not leave the EU. AI inference is also offered in Asia Pacific with regional processing, though voice agents are not offered there, and Canada has neither. The assistant's language model can also point at the customer's own OpenAI compatible endpoint, such as one on AWS Bedrock, Azure OpenAI or a self run vLLM server, which keeps inference inside an environment the customer controls.

Sourcetelnyx.com/ai/compliance.mdread 2026-10-11

The studio offers templates alongside plain English building, and Plivo's use cases cover lead capture, customer support, appointment booking, tech support, surveys and multichannel engagement. The Vibe Agent drafts a complete flow from a description, a fast start the team then shapes, as in Plivo's own example of sending a WhatsApp alert whenever a new Zendesk ticket arrives. Plivo sells no named agents a customer simply switches on.

Sourceplivo.com/ai-agentsread 2026-10-11

New assistants start from a template in the portal, with a blank template as the usual starting point, and Telnyx managed insights come ready to attach to any assistant. A shared tools library lets one set of tools serve many assistants. Telnyx offers no catalog of finished agents for support, sales or scheduling.

Sourcedevelopers.telnyx.com/docs/inference/ai-assistants/no-code-voice-assistant.mdread 2026-10-11

Each AI conversation node carries its own model setting, so the builder picks the LLM step by step, with GPT-4.1 as one option and Plivo advising teams to try several models to balance answer quality against latency. The studio's models come from the GPT and Llama families, with voices from Cartesia and ElevenLabs. In the Playground, analysis mode reports metrics for each model run. Developers on Plivo's code path pick their own speech to text, LLM and text to speech providers, such as Deepgram Nova 3 and Cartesia Sonic 2.

Sourcedocs.plivo.com/docs/aiagent/aistudio/nodereference/aiconversationread 2026-10-11

The customer picks the model for each assistant on the Agent tab. Options include models hosted on Telnyx GPUs, with Kimi as the default, managed third party models from providers such as OpenAI, Gemini and Groq, with OpenAI models on the customer's own key, and any public OpenAI compatible endpoint such as AWS Bedrock, Azure OpenAI, Baseten, vLLM or SGLang. Prompt nodes in a workflow can override the model step by step, and speech to text and text to speech providers are chosen separately.

Sourcedevelopers.telnyx.com/docs/inference/ai-assistants/custom-llm.mdread 2026-10-11

Any flow can expose an API trigger that outside systems call over HTTP with basic auth and input parameters, starting the agent programmatically, so a CRM or scheduler can have an agent place an outbound call or send a message. Around the studio sit Plivo's voice and messaging APIs, SDKs and a CLI, and audio streaming and SIP trunking interfaces for agents built on other frameworks.

Sourceplivo.com/docs/aiagent/aistudio/triggerread 2026-10-11

The API covers assistants end to end, including creating and updating assistants, assistant tests, canary deploys, MCP server records, SMS chat and a chat endpoint in beta, and the TeXML ai_calls endpoint starts outbound calls. Coding assistants can read the whole reference through llms.txt indexes, and the conversations API hands stored conversations to other systems.

Sourcedevelopers.telnyx.com/development/llms/ai-assistants-llms-txtread 2026-10-11

The Playground runs a flow before launch from a simulated inbound hello, with the trigger variables entered by hand. Transcript mode plays the conversation close to the live channel, and analysis mode shows each node's prompt, tool response and timing along with model metrics such as response time and accuracy. Test runs in the Playground are not billed. Simulation testing is included in the agent price, though Plivo does not say what it adds beyond the Playground or whether test cases can be saved and run again.

Sourceplivo.com/docs/aiagent/aistudio/playgroundread 2026-10-11

AI tests give each check a name, a target assistant and success criteria, such as what the greeting must say, and a test run shows progress live and reports whether the assistant met every criterion, with the conversation to review. A new version can be saved with notes and sent a share of traffic by ordered rules on the caller's number, all to one version or split by percentage, with the remainder staying on the main version, and rollback clears every rule at once. Coval can generate test conversations from seed cases.

Sourcedevelopers.telnyx.com/docs/inference/ai-assistants/version-testing-traffic-distribution.mdread 2026-10-11

Agents act through app actions, HTTP requests, Python code nodes and call forwarding, and the knowledge base crawls websites when content is synced. No hosted browser, remote desktop or screen control is part of the studio, so a system without an API is out of reach unless the team writes code to call it.

Sourceplivo.com/docs/aiagent/getstarted/overviewread 2026-10-11

Assistants act through webhooks, MCP servers, integrations and call control tools, and send keypad tones to other phone systems. No hosted browser, remote desktop or screen control is offered.

Sourcedevelopers.telnyx.com/docs/inference/ai-assistants/tools-library.mdread 2026-10-11