DigitalOcean Launches Inference Router for Automated Model Selection
DigitalOcean has introduced the Inference Router in Public Preview, which automatically routes API requests to the optimal model based on task-specific benchmarks. This feature eliminates the need for manual routing rules and redeployment when swapping between frontier and open-source models.
Verified State Diff
Impact & Verification Analysis
Developers and AI engineers using DigitalOcean's AI model catalog for production inference.
It reduces operational overhead by automating model selection and evaluation, potentially lowering costs by approximately 22% compared to using single frontier models for all tasks.
Full Fact Overview
The DigitalOcean Inference Router automates model selection by evaluating incoming requests against a catalog of frontier and open-source models (including Anthropic, OpenAI, Kimi, GLM, and DeepSeek). The system uses internal benchmarks to match tasks—specifically software engineering, agentic workflows, and general reasoning—to the most efficient model. It includes cache-aware routing to minimize latency and cost, and allows users to set a 'routing_preference' (balanced, budget, or quality) to influence model selection. The service is currently in Public Preview and accessible via the 'inference-router' model identifier in the API.