New playbook: when your own AI model pays. Read
Targa Telematics (connected mobility) reported this on 14 Oct 2026 in company self-report (source). It belongs to the models and routing layer.
An on-premise generative-AI development environment on HPE hardware (with MSP Gruppo Zenit) running open-weight LLMs for coding, so no code or data leaves company infrastructure.
First seven weeks: ~30,000 developer requests, 3.4B+ tokens processed, 65M tokens generated, ~150 engineers using it.
The number covers adoption/volume only; cost or quality comparison published is not reported.
One gateway picks the AI model for each task and records what it costs.
Up to 90%
lower AI cost, AT&T says
An internal AI Gateway that, turn by turn, weighs speed, cost and expected quality and routes each prompt to the most cost-effective model, switching models mid-session, plus a library of post-trained open telco models (OTel 2.0) that run on-prem or in cloud (source).
<8%
of closed-model cost per transaction
A blended model portfolio (in-house compact models, open-source models post-trained on Pinterest data in its own secure cloud, closed models only in 'very limited cases') behind a centralized routing layer that sends routine work to lighter models; being extended to internal engineering and sales tools in H2 2026 (source).
52%
lower cost per agent session since the June peak
A cost equation and weekly metric stack for its agentic 'software factory' (70%+ of PRs agent-attributed), with benchmark-driven Pareto model selection (incl. open-weight models), cheaper default subagent models, Medium reasoning effort default, 400K-token compaction cap and prompt-cache strategy (source).