--- title: Resources - Telnyx Blog - inference canonical: https://telnyx.com/fil-ph/resources/topic/inference --- # Resources - Telnyx Blog - inference Resources # Read it, clone it, or ask the machine. Infrastructure for real-time agents, documented as it ships. Release notes as features land, guides from the engineers who built the network, builds you can clone, and machine-readable versions of everything. [START BUILDING](https://portal.telnyx.com/#/login/sign-up) [RELEASE NOTES](#releases) [DOCS](https://developers.telnyx.com/docs/overview) Shipped this month * [28 OctPipecat Voice Agents Now Run on Telnyx Edge](https://telnyx.com/release-notes/pipecat-voice-agents-telnyx-edge) * [28 OctEmail Domains, Templates, and Validation Now Available in the Telnyx Portal](https://telnyx.com/release-notes/email-portal-domains-templates-validation) * [27 OctSMTP Relay Now Available for Email API](https://telnyx.com/release-notes/email-smtp-relay) * [26 OctMM7 Gateway Now Available for Telnyx MMS](https://telnyx.com/release-notes/mm7-mms-gateway) * [23 OctDedicated IP Addresses Now Available for Email API](https://telnyx.com/release-notes/email-dedicated-ip) * [22 OctAgent Browser Now Available for Telnyx AI Agents](https://telnyx.com/release-notes/agent-browser-now-available) * [21 OctCaller Identity and Enhanced Caller Name Now Available in Telnyx Number Lookup](https://telnyx.com/release-notes/number-lookup-caller-identity-enhanced-caller-name) Open source builds 517 Releases in 30 days 12 Countries covered 140+ Companies building 14,000+ Start here ## Four doors, by what you’re doing. * [What shippedRelease notes as features land, dated and tagged by product.Release notes](#releases) * [Building todayDocs start with a curl. Typed reference, SDKs in seven languages.developers.telnyx.com](https://developers.telnyx.com/docs/overview) * [CloningOpen-source agent builds on GitHub. Working code, README, .env.example.github.com/team-telnyx](https://github.com/team-telnyx) * [For machinesEvery surface above as text and JSON, for the agents reading this.Machine-readable](#machines) Release notes ## What shipped. Dated, by product, as it happens. Weekly or faster. Pick a product to see everything in that line. Filter by product All12 * [28 Oct 2026Pipecat Voice Agents Now Run on Telnyx EdgeDeploy Pipecat voice agents to Telnyx Edge with one command and connect them to a Telnyx phone number.Read the note](https://telnyx.com/release-notes/pipecat-voice-agents-telnyx-edge) * [28 Oct 2026Email Domains, Templates, and Validation Now Available in the Telnyx PortalSet up sending domains, build reusable templates, and validate recipient addresses directly in the Telnyx portal.Read the note](https://telnyx.com/release-notes/email-portal-domains-templates-validation) * [27 Oct 2026SMTP Relay Now Available for Email APISend email through the Telnyx SMTP relay with the same domains, quotas, billing, and events as API sends.Read the note](https://telnyx.com/release-notes/email-smtp-relay) * [26 Oct 2026MM7 Gateway Now Available for Telnyx MMSSend and receive MMS over the MM7 protocol, the standard SOAP/XML interface used by carrier MMSCs and enterprise MMS platforms.Read the note](https://telnyx.com/release-notes/mm7-mms-gateway) * [23 Oct 2026Dedicated IP Addresses Now Available for Email APIDedicated sending IPs for high-volume email customers.Read the note](https://telnyx.com/release-notes/email-dedicated-ip) * [22 Oct 2026Agent Browser Now Available for Telnyx AI AgentsGive an AI agent its own isolated browser session through an API, limited to the sites you allow, so it can use websites that have no API.Read the note](https://telnyx.com/release-notes/agent-browser-now-available) * [21 Oct 2026Caller Identity and Enhanced Caller Name Now Available in Telnyx Number LookupTelnyx Number Lookup now returns the name and postal address behind a phone number with Caller Identity, plus Enhanced Caller Name with broader coverage than standard CNAM.Read the note](https://telnyx.com/release-notes/number-lookup-caller-identity-enhanced-caller-name) * [20 Oct 2026Prompt and Keyterm Support Now Available for AssemblyAI on Telnyx Voice AI AssistantsAssemblyAI Universal-3.5 Pro on Telnyx Voice AI Assistants now accepts a context prompt and keyterms, and receives the agent's last reply on every turn.Read the note](https://telnyx.com/release-notes/assemblyai-prompt-keyterm-voice-ai-assistants) Showing the 8 most recent of 12 notes. [View all releases](https://telnyx.com/release-notes) Guides and articles ## Written by the people who run the network. Not sure where to start? Browse the library by product - deep dives, setup guides and working demos, every one dated, bylined and never behind a form. [All211](https://telnyx.com/fil-ph/resources#learn)[Conversational AI45](https://telnyx.com/fil-ph/resources/topic/conversational-ai#learn)[Storage3](https://telnyx.com/fil-ph/resources/topic/storage#learn)[Wireless10](https://telnyx.com/fil-ph/resources/topic/wireless#learn)[SIP Trunking14](https://telnyx.com/fil-ph/resources/topic/sip-trunking#learn)[Regulations2](https://telnyx.com/fil-ph/resources/topic/regulations#learn)[Product1](https://telnyx.com/fil-ph/resources/topic/product#learn)Inference36[Numbers4](https://telnyx.com/fil-ph/resources/topic/numbers#learn)[Networking5](https://telnyx.com/fil-ph/resources/topic/networking#learn)[Voice16](https://telnyx.com/fil-ph/resources/topic/voice#learn)[Integrations2](https://telnyx.com/fil-ph/resources/topic/integrations#learn)[IoT4](https://telnyx.com/fil-ph/resources/topic/iot#learn)[eSIM2](https://telnyx.com/fil-ph/resources/topic/e-sim#learn)[Verify API2](https://telnyx.com/fil-ph/resources/topic/verify-api#learn)[Fax4](https://telnyx.com/fil-ph/resources/topic/fax#learn)[General3](https://telnyx.com/fil-ph/resources/topic/general#learn)[Telnyx1](https://telnyx.com/fil-ph/resources/topic/telnyx#learn)[Messaging17](https://telnyx.com/fil-ph/resources/topic/messaging#learn)[Email API4](https://telnyx.com/fil-ph/resources/topic/email-api#learn)[WhatsApp Business API7](https://telnyx.com/fil-ph/resources/topic/whatsapp-business-api#learn) Search articles * [![Laptop downloading open weight model files](https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/open-weight-models-1920x1080.svg)InferenceOpen weight models: what they are and how to use themEli Mogul · 23 Oct 2026](https://telnyx.com/resources/open-weight-models) * [![Laptop chat reply beside an inference GPU rack](https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/inference-platforms-1920x1080.svg)Inference Top AI inference platforms: how to evaluate providers on real costEli Mogul · 22 Oct 2026](https://telnyx.com/resources/inference-provider-eval) * [![Inference as a service guide cover](https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/inference-as-a-service-1920x1080.svg)InferenceInference as a service: how it works and what it costsEli Mogul · 21 Oct 2026](https://telnyx.com/resources/inference-as-a-service) * [![GPU card beside phone handset in rack](https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/ai-inference-hardware-1920x1080.svg)InferenceAI inference hardware: How to choose chips and placementEli Mogul · 20 Oct 2026](https://telnyx.com/resources/ai-inference-hardware) * [![Multi-turn LLM benchmark featured image](https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/featured-multi-turn-llm-benchmark-3-models-4-providers.png)InferenceMulti-turn LLM benchmark: 3 models, 4 providersSonam Gupta · 20 Oct 2026](https://telnyx.com/resources/multi-turn-llm-benchmark) * [![GPU card, stopwatch and streaming chat reply](https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/inference-optimization-1920x1080.svg)InferenceInference optimization: cut LLM latency and costEli Mogul · 16 Oct 2026](https://telnyx.com/resources/inference-optimization) * [![GLM 5.3 Flash benchmark: Telnyx vs Modal](https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/glm53-flash-benchmark-featured.png)InferenceGLM 5.3 Flash benchmark: Telnyx vs ModalSonam Gupta, PhD · 15 Oct 2026](https://telnyx.com/resources/glm53-flash-benchmark) * [![RT voice inference feature](https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/real-time-voice-inference-1920x1080.svg)InferenceReal-time voice inference: where every millisecond goes on a live callEli Mogul · 15 Oct 2026](https://telnyx.com/resources/real-time-voice-inference) * [![Edge inference feature](https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/edge-inference-featured-1920x1080-1.svg)InferenceEdge inference: where to run AI models for low latencyEli Mogul · 14 Oct 2026](https://telnyx.com/resources/edge-inference) * [![Open source vs frontier feature](https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/open-source-vs-frontier-models-1920x1080.svg)InferenceOpen Source vs Frontier Models: Which Should Your Agents Run On?Eli Mogul · 14 Oct 2026](https://telnyx.com/resources/os-vs-frontier) * [![NVIDIA B300 pricing per hour and system](https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/b300-pricing-featured-1920x1080.svg)InferenceNVIDIA B300 pricing in 2026: per hour, per system and per tokenEli Mogul · 8 Oct 2026](https://telnyx.com/resources/nvidia-b300-pricing) * [![GLM-5.3 Flash, GLM-5.3, and Kimi K3 across 45 multi-turn agent conversations](https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/multi-turn-llm-benchmark.png)InferenceWe tested 3 open-weight models for multi-turn agentsSonam Gupta, PhD · 8 Oct 2026](https://telnyx.com/resources/choose-open-weight-model) * [![Inference infrastructure feature image](https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/ai-inference-infrastructure-1920x1080.svg)InferenceInference Infrastructure: What to Run Where and Why It MattersEli Mogul · 8 Oct 2026](https://telnyx.com/resources/inference-infrastructure) * [![Inference engine feature image](https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/01-faithful-1920x1080.svg)InferenceWhat is an inference engine? Types, uses, and vLLMEli Mogul · 6 Oct 2026](https://telnyx.com/resources/inference-engine) * [![](https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/58db2ca1-0d1d-40c8-9ef5-a79897cb9ce9-a2f51d62.jpeg)InferenceOpen-Source Models Are Catching Up to FrontierFiona McDonnell · 6 Oct 2026](https://telnyx.com/resources/open-source-models-catching-up-to-frontier) * [![Avoiding indirect prompt injection in AI coding agent instruction files like AGENTS.md and CLAUDE.md](https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/ai-prompt-injection-avoidance.svg)InferenceHow to scan AGENTS.md and CLAUDE.md for indirect prompt injectionSonam Gupta, PhD · 5 Oct 2026](https://telnyx.com/resources/scan-agents-md-prompt-injection) * [![DLP Test for AI Gateway](https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/AI%20DLP%20Test%20Charts%20v2-selection%20small.png)InferenceAI DLP Test: Six Data Types, Six Blocks, $0 ChargedFiona McDonnell · 2 Oct 2026](https://telnyx.com/resources/ai-dlp-gateway-test) * [![](https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/58db2ca1-0d1d-40c8-9ef5-a79897cb9ce9-a2f51d62.jpeg)InferenceInference Cost Optimization: How to Cut Your AI Bill by up to 75%Fiona McDonnell · 18 Sep 2026](https://telnyx.com/resources/inference-cost-optimization) * [![](https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/58db2ca1-0d1d-40c8-9ef5-a79897cb9ce9-a2f51d62.jpeg)InferenceGLM-5.3 latency benchmarks across four inference providersSonam Gupta, PhD · 18 Sep 2026](https://telnyx.com/resources/glm-5-3-latency-benchmarks-provider) * [![Edge site density feature image](https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/feature-image-edge-site-density_1---b6536fca-5fc8-498d-bfec-2f790c55b135-1.svg)InferenceWhy Edge Site Density MattersMegha Sujanani · 18 Sep 2026](https://telnyx.com/resources/why-edge-site-density-matters) * [![](https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/58db2ca1-0d1d-40c8-9ef5-a79897cb9ce9-a2f51d62.jpeg)InferenceWhat is serverless AITiffany McDowell · 17 Sep 2026](https://telnyx.com/resources/serverless-ai-explained) * [![TPU vs GPU feature image](https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/05-dual-chip-1920x1080.svg)InferenceTPU vs GPU Compared for AI Training and InferenceEli Mogul · 15 Sep 2026](https://telnyx.com/resources/tpu-vs-gpu) * [![Choosing the right storage solution for AI](https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/M2_M_SIM_card_308fde4d84.png)InferenceChoosing the right storage solution for AIKelsie Anderson · 10 Sep 2026](https://telnyx.com/resources/storage-for-ai) * [![](https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/58db2ca1-0d1d-40c8-9ef5-a79897cb9ce9-a2f51d62.jpeg)InferenceBenefits and challenges of using embeddings databasesKelsie Anderson · 10 Sep 2026](https://telnyx.com/resources/embeddings-database) * [![how-to-scale-ai-cost](https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/how_to_scale_ai_cost_featured_7f230286b4.webp)InferenceAI on demand: How to scale with serverless efficiencyTiffany McDowell · 10 Sep 2026](https://telnyx.com/resources/how-to-scale-ai-cost) * [![Kimi K3 now available on Telnyx Inference](https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/c35a41bd-7d5b-4e0c-a3f6-0734a8d2badf.jpg)InferenceKimi K3, live on Telnyx InferenceFiona McDonnell · 10 Sep 2026](https://telnyx.com/resources/kimi-k3-open-model-telnyx-inference) * [![GLM-5.3-Flash blog header](https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/glm-5-3-flash-blog-header-1600x900_2---79980733-6619-43fe-b8c9-b91d23507a67.svg)InferenceGLM-5.3-Flash BenchmarksSonam Gupta, PhD · 10 Sep 2026](https://telnyx.com/resources/glm-5-3-flash-benchmarks) * [![](https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/58db2ca1-0d1d-40c8-9ef5-a79897cb9ce9-a2f51d62.jpeg)InferenceGLM-5.2 is available on Telnyx InfrastructureFiona McDonnell · 10 Sep 2026](https://telnyx.com/resources/glm-5-2-on-telnyx-inference) * [![Frontier models feature image](https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/01-faithful.svg)InferenceWhat are frontier models and why data sovereignty mattersEli Mogul · 10 Sep 2026](https://telnyx.com/resources/frontier-models) * [![DeepSeek V4 Flash inference benchmark data](https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/deepseek-v4-flash-header.svg)InferenceDeepSeek V4 Flash inference benchmarks on TelnyxSonam Gupta, PhD · 10 Sep 2026](https://telnyx.com/resources/deepseek-v4-flash-benchmarks) Page 1 of 2. [Next](https://telnyx.com/fil-ph/resources/topic/inference/page/2#learn) Compare ## Three layers. Three sets of competitors. Every page is built from the competitor’s public rate card, with the date it was pulled. Where they are excellent, we say so. ### Edge Compute 4 pages Cloudflare put compute at the edge of the internet. We put it at the edge of the phone network. Their network reaches the internet; ours reaches the internet and the phone network. * [Primary frameTelnyx vs CloudflareRead](https://telnyx.com/the-best-cloudflare-alternative) * [vsTelnyx vs Together.aiRead](https://telnyx.com/the-best-together-alternative) * [vsTelnyx vs FireworksRead](https://telnyx.com/the-best-fireworks-alternative) * [vsTelnyx vs BasetenRead](https://telnyx.com/the-best-baseten-alternative) ### Agent Platform 4 pages Prototypes assemble. Production owns. When the demo becomes a business, the rented stack is where it breaks. * [vsTelnyx vs VapiRead](https://telnyx.com/the-best-vapi-alternative) * [vsTelnyx vs RetellRead](https://telnyx.com/the-best-retell-alternative) * [vsTelnyx vs ElevenLabsRead](https://telnyx.com/the-best-elevenlabs-alternative) * [vsTelnyx vs LiveKitRead](https://telnyx.com/the-best-livekit-alternative) ### Global Communications 4 pages Telnyx is the carrier. We hold the licenses, own the numbers, and control the interconnects, other CPaaS providers rent numbers and buy PSTN access from someone who does. * [vsTelnyx vs TwilioRead](https://telnyx.com/the-better-twilio-alternative) * [vsTelnyx vs BandwidthRead](https://telnyx.com/the-better-bandwidth-alternative) * [vsTelnyx vs VonageRead](https://telnyx.com/the-better-vonage-voice-api-alternative) * [vsTelnyx vs Amazon ConnectRead](https://telnyx.com/amazon-connect-alternative-why-telnyx-is-a-smarter-choice) Also compared [Amazon S3](https://telnyx.com/the-better-amazon-s3-alternative) [Backblaze B2](https://telnyx.com/the-better-backblaze-b2-alternative) [Azure Blob](https://telnyx.com/the-better-microsoft-azure-blob-alternative) [Google Cloud Storage](https://telnyx.com/the-better-google-cloud-storage-alternative) [1NCE](https://telnyx.com/the-better-1nce-alternative) [Emnify](https://telnyx.com/the-better-emnify-alternative) [Hologram](https://telnyx.com/the-better-hologram-alternative) [Kore Wireless](https://telnyx.com/the-better-kore-wireless-alternative) Every comparison cites the competitor’s public card and the date pulled. Rates on [telnyx.com/pricing](https://telnyx.com/pricing). For the agents reading this ## Everything above, as data. Read the docs, the API surface, and the signup flow without a human in the loop. * [llms.txtIndex of every documented surface. Start here if you're a model.Text](https://telnyx.com/llms.txt) * [llms-full.txtThe full corpus in one file, for a single fetch into context.Text](https://telnyx.com/llms-full.txt) * [openapi.jsonThe complete v2 API surface as an OpenAPI document.JSON](https://telnyx.com/openapi.json) * [agent-signup.mdProvision an account programmatically. No sales queue.Markdown](https://telnyx.com/agent-signup.md) * [agents/startThe fast path: what to read, in what order, to start building.Text](https://telnyx.com/agents/start) ## Read less. Clone more. Working code and a curl beat a whitepaper. When you want a human, they run the network and pick up 24/7. [START BUILDING](https://portal.telnyx.com/#/login/sign-up) [BROWSE BUILDS](https://github.com/team-telnyx) [TALK TO AN ENGINEER](https://telnyx.com/contact-us) [Start building](https://telnyx.com/sign-up) ### Ask AI * [![GPT](https://d2efw2akbx8gaf.cloudfront.net/https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/chatgpt_icon-ea55e09d.jpeg?width=92&format=webp)](https://chat.openai.com/?q=What makes Telnyx's global network different, breakdown call quality for normal phone calls and voice AI calls.) * [![Claude](https://d2efw2akbx8gaf.cloudfront.net/https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/claude_icon-c7c8eef3.webp?width=92&format=webp)](https://claude.ai/new?q=What makes Telnyx's global network different, breakdown call quality for normal phone calls and voice AI calls.) * [![Perplexity](https://d2efw2akbx8gaf.cloudfront.net/https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/perplexity_icon-cd5b74f1.webp?width=92&format=webp)](https://www.perplexity.ai/search?q=What makes Telnyx's global network different, breakdown call quality for normal phone calls and voice AI calls.) * [![Gemini](https://d2efw2akbx8gaf.cloudfront.net/https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/gemini_icon-4ff1cbf4.webp?width=92&format=webp)](https://www.google.com/search?udm=50&aep=11&q=What makes Telnyx's global network different, breakdown call quality for normal phone calls and voice AI calls.) * [![Grok](https://d2efw2akbx8gaf.cloudfront.net/https://dotcom-payload-assets-prod.s3.us-east-1.amazonaws.com/cms-media/grok_icon-e588e5bd.webp?width=92&format=webp)](https://x.com/i/grok?text=What makes Telnyx's global network different, breakdown call quality for normal phone calls and voice AI calls.)