We’re seeing model routing deliver significant savings across large enterprises and all workflows. Full model independence and reliable harness-level routing enables this access to the Pareto frontier.
Excited to share this work with everyone! More to come. Expect more savings as efficient models and routing continue to improve
Router has been in production for a few weeks and the savings have exceeded the original benchmarks. Measuring billed usage vs the same workload priced at Opus rates, we’ve found:
43% aggregate cost savings
61% of sessions are at least 80% cheaper