New playbook: when your own AI model pays. Read
Databricks (data and ai software, ~9,000 employees) reported this on 22 Oct 2026 in company self-report (source). It belongs to the models and routing layer.
An internal control layer between developers and models: moving defaults to newer cheaper models in one place, task-aware routing across models and harnesses, per-developer spend reports, and progressive friction (warning, approval, downshift to cheaper model) in place of hard caps.
Unit costs down 'as much as 90%' in some cases; model switching alone 50%+; spend reports ~10%; Smart Router cut average task cost by more than 30% at roughly top-model quality (65% of Opus 5 cost per task on internal coding workloads); output close to doubled with the same team.
The number covers databricks' own AI-coding usage; vendor dogfooding (it sells the gateway).
“…in some cases, our unit costs are down as much as 90%,” Databricks said.
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).