Infrastructure for real-time agents.

A licensed carrier that owns the network, the edge compute and the GPUs. Run inference, deploy agents and reach people and machines on one platform.

Start building Explore Inference API

14,000+ INDUSTRY-LEADING COMPANIES choose telnyx

Inference platform

Inference infrastructure that scales with you

Run open-weight models serverlessly with no infrastructure to manage and no long-term commitments. Move to dedicated inference when you need predictable performance, greater control, and better economics at scale. Bring open-source, custom, or fine-tuned models, or reserve dedicated GPU capacity for your own workloads.

Try Inference API View pricing model

Keep your OpenAI SDK
Set the Telnyx base URL, API key, and model in your existing client.
Streaming and tool calling
Stream responses as they are generated. Supported models can also call tools and return structured outputs.
Global inference footprint
Run models closer to your users on Telnyx GPU infrastructure.
Serverless inference
Run open-weight models on demand with usage-based pricing, no infrastructure to manage, and no long-term commitments.
Dedicated inference
Run open-source, custom and fine-tuned models on GPU infrastructure built for high-throughput, low-latency inference.
GPU capacity
Reserve dedicated GPU capacity for your own workloads, including custom inference stacks and fine-tuned models.

Our models

GLM-5.3-Flash
Understands text and images, and delivers strong coding performance at a fraction of the compute
GLM-5.3
Built for complex software engineering, from multi-step agent runs to spotting vulnerabilities in code
GLM-5.2
Coding, reasoning, 1M context window
DeepSeek-V4.1-Flash
Reads images and a million tokens of context, with reasoning you can dial up or down to control cost
DeepSeek-V4-Flash-0731
Advanced coding, tool use, and long-horizon agentic workflows
telnyx-decision-flash
Lowest cost and latency for high-volume decisions
telnyx/decision-pro
Decisions that need long context, including inputs beyond Jev’s 32k per-decision limit
Kimi-K3
State-of-the-art open-weight intelligence for coding, reasoning, and multimodal work
Kimi-K2.6
Voice AI
MiniMax-M3-MXFP8
Lowest cost at high intelligence
Qwen3.8-27B
Multimodal coding and visual understanding across documents, diagrams, and video
Qwen3-Embedding-8B
Multilingual embedding model for search and RAG, with 32K context and up to 4,096 dimensions.

GLM-5.3 Published benchmark

Workloads with the lowest median completion time

Read the benchmark

Telnyx — Median full-response latency · approximately 88.7k input / 1,200 output tokens

Together — Median full-response latency · approximately 88.7k input / 1,200 output tokens

COMPOSITION

50 primitives. One control plane. Watch the same blocks become five different products.

Start with one product. Inference, voice, storage, or wireless. They sit on the same infrastructure and appear on the same bill, so adding a channel never adds a vendor. Choose a workload to see what it composes, or pick a product directly.

CHOOSE BY USE CASE OR PRODUCT

CommsGlobal numbers

RESULTS

cURL

https://telnyx.com/products/phone-numbers

Your stack

Primitives composed: 1 of 50
Pricing: Per-product pricing
Vendors required: 1

SELECT USE CASE

OR BUILD YOUR OWN STACK

RESULTS

cURL

https://telnyx.com/products/phone-numbers

Your stack

Primitives composed: 1 of 50
Pricing: Per-product pricing
Vendors required: 1

RESULTS

cURL

https://telnyx.com/products/phone-numbers

Your stack

Primitives composed: 1 of 50
Pricing: Per-product pricing
Vendors required: 1

Runtime

Swap the cloud edge for the telco edge

Run your code in containers at Telnyx points of presence (PoPs). Persistent state carries context between sessions, while your application calls the communications and inference APIs it needs.

Compute/docs/edge-compute/overview
Functions
Deploy containerized applications with your own dependencies at Telnyx points of presence.
Run it
Compute/docs/edge-compute/stateful-actors
StatefulActor
Each entity has a persistent instance whose state survives restarts. Calls run one at a time.
Run it
Storage & Data/docs/cloud-storage/overview
Object Storage
S3-compatible storage for your application data, including recordings and model artifacts.
Run it
Storage & Data/docs/edge-compute/kv
KV
Store session context with a TTL so agents can read it across channels.
Run it
Storage & Data/docs/edge-compute/sqldb
SQLDB
Serverless SQLite for edge functions. Relational state without a database to run.
Run it
Storage & Data/docs/edge-compute/cloudfs
CloudFS
Access files in object storage through a filesystem mount.
Run it
AI & SDK/docs/inference/getting-started
Inference API
Call open models from your edge application, in region. Stream responses and use tool calling through an OpenAI-compatible API.
Run it
AI & SDK/docs/agent-sdk
Agent SDK
Build agents with persistent state and memory, with scheduled tasks and bindings to Telnyx communications APIs.
Run it

Infrastructure

We operate the carrier network and the GPUs

Telnyx is a licensed carrier with its own private global backbone. We own the edge compute and GPUs, so the network and model infrastructure are available from the same provider, under one contract and one compliance boundary.

Start building

Countries: 140+; With numbers and voice coverage.
Points of presence: 25+; On a private backbone.
Primitives: 50; On one control plane.
Languages: 100+; In real time.

Ownership

We own every layer an agent touches

Ownership is the load-bearing claim. Price, latency, reliability, identity, and sovereignty are consequences of it. A competitor can match a price. They cannot buy the structure that produced it.

Edge compute

Functions, KV, Stateful Actors, and Object Storage at the PoP. Inference on owned GPUs, colocated with the media plane.

Agent platform

Voice AI, STT, TTS, and orchestration on one control plane. Primitives that compose into agents, contact centres, and connected fleets.

Global comms

Voice, messaging, and numbering on a licensed carrier network. We hold the telecom licence, not a reseller arrangement.

NETWORK

Explore our PoPs

Filter site types

AI & Compute

Communications

Connectivity

Sovereignity & Compliance

Competitors

What you get from each platform

Each platform covers a different part of the application. Telnyx combines carrier services with model serving and an application runtime.

LAYER

Licensed carrier / PSTN

Own

Rent

Absent

Rent

Private global network

Own

Absent

Own

Absent

Edge PoPs

Own

Rent

Own

Absent

GPUs & Inference

Own

Absent

Own

Rent

Wireless & SIM / mobile core

Own

Absent

Programmatic compliance (KYC, 10DLC, e911)

Own

Absent

Rent

Agent orchestration

Own

Rent

Own

Complete voice AI turn, one vendor

Complete

Partial

No PSTN

Assembled

PRICING

Price your Voice AI workload

Use your call volume to estimate voice AI costs. For model usage, open the Inference API calculator.

Conversations per month

01M

Average mins per conversation

015

SMS follow-ups per conversation

Voice + telephony: $0.006
STT + TTS + Inference (co-located): $0.150
Messaging: $0.004
Per conversation, all-in: $0.160

Telnyx per month: $76,800
Rented stack per month: ≈ $261,120

You keep $184,320/mo. That's the margin you weren't paying anyone for.

*Illustrative rates for mockup. Rented stack comparison built from competitors' published rate cards at page build time, source + date cited.

For the agents reading this

Infrastructure agents can discover, price, and buy themselves. Live now.

Every claim has a data twin: capability, coverage, and pricing, all versioned and signed. Provisioning is an API, not a sales queue.

llms.txt + /ai/pricing.jsonVersioned, live

// canonical machine claims
GET telnyx.com/llms.txt
  // index, constraints,
  // bias disclosure
GET telnyx.com/ai/pricing.json
{
 "schema_version": "1.0.0",
 "economics": {
  "voice_ai_agent_usd_per_min": 0.05,
  "sip_outbound_usd_per_min": 0.005,
  "sms_outbound_usd_per_msg": 0.004,
  "number_monthly_usd": 1
 }
}

MCP serverapi.telnyx.com/v2/mcp, live

// query the stack,
// don't read about it
POST api.telnyx.com/v2/mcp
  Accept: application/json,
    text/event-stream

{"jsonrpc":"2.0","id":1,
 "method":"tools/list"}
→ list_api_endpoints()
→ get_api_endpoint_schema(id)
→ invoke_api_endpoint(id, args)
// 3 tools reach the whole Telnyx API

Procurementagent-signup.md, live

// signup is an API,
// not a sales queue
GET telnyx.com/.well-known/
    agent-access.json
GET telnyx.com/agent-signup.md
POST api.telnyx.com/v2/bot_challenge
  → a problem only an LLM can solve
  → account, inbox, API key
    — no human in the loop
// no account at all:
// x402.telnyx.com, USDC per call

Sign up and start building

Everything an agent needs to evaluate and provision Telnyx is machine-readable and live. Docs, pricing and the full API over MCP, then an account and a key with no human in the loop.

AGENTS: telnyx.com/llms.txt

MCP: api.telnyx.com/v2/mcp

Infrastructure for real-time agents.

Inference infrastructure that scales with you

Keep your OpenAI SDK

Streaming and tool calling

Global inference footprint

Serverless inference

Dedicated inference

GPU capacity

Our models

Workloads with the lowest median completion time

Swap the cloud edge for the telco edge

Functions

StatefulActor

Object Storage

KV

SQLDB

CloudFS

Inference API

Agent SDK

We operate the carrier network and the GPUs

We own every layer an agent touches

Edge compute

Agent platform

Global comms

Explore our PoPs

Filter site types

What you get from each platform

Price your Voice AI workload

Infrastructure agents can discover, price, and buy themselves. Live now.

Sign up and start building

Ask AI