Industry Newssmall language modelsagentstelecomcost optimization
AT&T Cuts Assistant Costs With Small Models
8.2
AT&T boosted the efficiency of its internal Ask AT&T personal assistant by reworking the orchestration layer and shifting more work from large language models to small language models, VentureBeat reported Thursday (Feb. 26). The change lowered latency and response times, cut costs by about 90% and enabled the system to process three times as many tokens. The move indicates enterprises can use SLMs to reduce operational costs while reserving LLMs for high-stakes steps.
Key Points
- 1Shifted orchestration to small language models, improving latency and tripling token throughput for Ask AT&T.
- 2Cut costs by about 90%, demonstrating SLMs' cost-efficiency for domain-specific agent workflows.
- 3Enables reserving LLMs for rare high-stakes steps, reducing infrastructure and operational burdens.
Scoring Rationale
Demonstrates large enterprise cost and throughput gains, but evidence is from a single-company report source.
Sources
Public references used for this report.
Practice with real Telecom & ISP data
90 SQL & Python problems · 15 industry datasets
Used by DS/ML engineers at top companies
Active Residential CustomersEasyUnlimited Fiber Plans 500Mbps+MediumCustomer Churn Risk AssessmentHard
250 free problems · No credit card
See all Telecom & ISP problems
