Poetiq says its system optimized an LLM router in 12 hours
The test used simulated company workloads; Poetiq says its custom routers improved cost and accuracy against OpenRouter and TypeSafe's JEV on its benchmark.
By Ryan Merket · Published
Primary source: X
Why it matters
Poetiq is pitching automated orchestration as a way to improve model cost and accuracy without retraining models. Its results are company-reported and based on simulated workloads, so the commercial test is whether those gains hold on customers' real query mixes.
Poetiq says its Metasystem optimized a custom large language model router in 12 hours, producing a series of configurations that it says beat OpenRouter's Auto Router and TypeSafe's JEV on its own tests. The company says one engineer spent 2.5 weeks on the entire project, from gathering data through final results; the 12 hours cover the system's optimization work, not the full project.