Release notes

Frontier-class reasoning model with intelligence comparable to Kimi K3, at roughly half the cost and twice the speed. Hosted on Telnyx-owned GPU infrastructure.

Recipients of a transfer can verbally accept or decline before the caller is bridged, with a private consult step between the assistant and destination.

GLM-5.3-Flash is now available on the Telnyx Inference API, delivering GLM-5.2-class benchmark performance at a fraction of the cost, hosted on Telnyx-owned GPUs.

You can now register RCS brands and agents programmatically, from brand verification through carrier launch, via the Telnyx API.

You can now serve Edge Compute Functions from your own domain with your own TLS certificate.

Customers can submit, vote on, and track feature requests through the Telnyx portal with duplicate detection, screenshot attachments, and a public API.

Two new smaller, faster LLMs are now available on the Telnyx Inference API, DeepSeek-V4-Flash for cost-efficient workloads and Qwen3.8-27B for low-latency tasks, both served from Telnyx-owned GPU...

Declare per-key request limits in telnyx.toml and check them with a single function call. No external service, no KV namespace, no counter logic. Platform-managed fixed-window rate limiting for...

Web search, content retrieval, and cited research through one Telnyx-owned API. $5 per 1,000 calls. Uses your existing Telnyx API key.

AgentSDK provides a TypeScript base class for building AI agents on Edge Compute with persistent message history, durable scheduled tasks, and merge-patch state. Bring your framework unchanged:...

Cloud Storage is now available in ca-central-1 (Canada) with full feature parity to US and APAC regions, including knowledge bases, presigned URLs, public buckets, and SSL certificates.

Three new LLMs for Voice AI Assistants: GLM-5.2 for maximum intelligence, GPT-5.6 Luna for cost-efficient voice workloads, and GPT-5.6 Sol for deep reasoning.