About Telnyx
Telnyx provides telecom and AI infrastructure through APIs. Its platform covers model inference, agents, and reaching people and machines by phone and messaging.
What it does
Telnyx offers inference APIs for running open-weight and custom models without managing servers. Teams can choose serverless pricing or dedicated GPU capacity. Existing OpenAI SDK code can often be pointed at Telnyx by changing the base URL, the API key and the model name.
Key features
- Use an OpenAI SDK by changing the base URL
- Streaming responses and tool calling
- Serverless inference with usage-based pricing
- Dedicated GPU capacity for custom models
- Telecom APIs to reach people and machines
Who it's for
Developers building AI apps and voice agents that need model hosting and telephony in one place. Backend teams that already use OpenAI-compatible code and want to test open-weight models may also find it useful.
Pricing and good to know
Serverless inference is billed by usage, and dedicated capacity is available when you need steady performance. Pricing details are on the company's pricing page. Models on the platform differ in cost and strengths, so check the model list before you build on one.
Highlights
- Use an OpenAI SDK by changing the base URL
- Streaming responses and tool calling
- Serverless inference billed by usage
- Dedicated GPU capacity for custom models
- Telecom APIs to reach people and machines
