Vercel Adds GPT-5.6 Sol, Terra, and Luna to AI Gateway
Quick Read
Vercel has added GPT-5.6 from OpenAI to its AI Gateway platform in a limited preview, offering three model tiers, Sol, Terra, and Luna, with no platform markup on inference.
OpenAI’s latest model generation just became a lot more accessible for developers building on Vercel.
GPT-5.6 from OpenAI is now available on Vercel’s AI Gateway in a limited preview, across three models: Sol, Terra, and Luna. All three perform better at agentic work across coding, biology, and cybersecurity, and are more token-efficient than the previous generation.
Each model serves a different use case. Sol is the flagship and most capable of the three. Terra is a balanced model for everyday work, with performance comparable to the previous generation at half the cost. Luna is the fastest and most affordable option, offering strong capability at the lowest cost in the series. Together, they give engineering teams meaningful flexibility depending on the workload and budget.
Getting started requires minimal code change. Developers can set the model to one of the GPT-5.6 slugs in the AI SDK. Alternatively, routing rules allow teams to switch from other gateway models to GPT-5.6 without touching their codebase. That flexibility is particularly useful for teams already running GPT-5.5 workloads who want to upgrade quietly and gradually.
Vercel’s AI Gateway provides a unified API for calling models, tracking usage and cost, and configuring retries, failover, and performance optimisations for higher-than-provider uptime. It includes built-in custom reporting, Zero Data Retention support, budgets for API keys, and routing rules. Notably, the platform reflects provider pricing with no markup and does not charge a platform fee on inference, including on Bring Your Own Key (BYOK) requests.
For developers across Nigeria and Africa building AI-powered products, this is a meaningful opening. GPT-5.6 AI Gateway access through Vercel removes one layer of cost and complexity that has historically made frontier model deployment harder for smaller teams. With zero markup and BYOK support, cost predictability improves significantly.
Teams can try GPT-5.6 directly in the model playground on AI Gateway. The leaderboard also tracks the most popular models by token volume across all Gateway traffic, useful for benchmarking what others are running at scale.
