DigitalOcean Serverless Inference Now Serves OpenAI GPT-6 Models

Melvin Vivas · X video post · 2026-09-24 · Open on X

Topics: LLMOps, Deployment & Monitoring, LLM Fundamentals · Level: intermediate

Summary

DigitalOcean now offers OpenAI models through its Serverless Inference product. The post suggests GPT-6 Luna for high-volume classification and routing, and GPT-6 Sol for multi-step reasoning and coding. Billing is combined with the agents that call the models.

Key points

• DigitalOcean now works as an AI model provider through Serverless Inference.
• GPT-6 Luna: meant for classification and routing at high volume.
• GPT-6 Sol: meant for multi-step reasoning and coding.
• Model usage goes on the same bill as the agents that call the models.
• The models can be tried in DigitalOcean's Inference Engine and Model Library.
• Model routing pattern: send cheap, high-volume tasks to a small model and hard reasoning to a stronger one.

Resources mentioned

• DigitalOcean Model Library · website · digitalocean.com · check price
  DigitalOcean's catalog for comparing and using large language models available through its inference service.
• DigitalOcean Serverless Inference · tool · digitalocean.com · paid
  DigitalOcean's managed serverless API for calling hosted LLMs.
• GPT-6 Luna · tool · community.openai.com · paid
  An OpenAI model aimed at fast, high-volume classification and routing.
  Also in: Codex Orchestrator v0.2.1: GPT-6 Sol orchestrator with Luna subagents (Melvin Vivas on X · notes), DeepSWE results: GPT-6 Sol slightly below GPT-5.6 Sol, but cheaper (Melvin Vivas on X · notes), OpenAI Releases GPT-6 Sol and GPT-6 Luna (Melvin Vivas on X · notes)
• Luna · tool · openai.com · paid
  The other model/agent the classifier routes tasks to; the post doesn't describe it further.
  Also in: GPT-6.1 Sol May Beat Luna for Subagents (Melvin Vivas on X · notes), Picking models for orchestrator and subagent roles in Codex (Melvin Vivas on X · notes), GPT-6 Sol Ultra Subagents Use Up Limits Fast (Melvin Vivas on X · notes), GPT-6 Sol vs Opus 5.5 in a Livestream Comparison (Melvin Vivas on X · notes) and 14 more

Try this

• Try GPT-6 Luna and Sol in DigitalOcean's Inference Engine.
• Build a router that sends simple classification jobs to GPT-6 Luna and complex reasoning jobs to GPT-6 Sol.