GPT-5.6 Sol is now 50% off a lower price
GPT-5.6 Sol now costs less across Vercel AI Gateway tiers, with an additional 50% discount through September 18 and no model-ID change for existing agent setups.
GPT-5.6 Sol list pricing fell **20% for input** and **one-third for output**. Vercel's 50% AI Gateway discount applies to the lower rates through **September 18** across Default, Flex, and Priority tiers.
Existing requests keep the same model ID and receive the new rate automatically. For agent workloads up to **272K tokens**, recheck cost assumptions for long investigations and high-output coding runs.
GPT-5.6 Sol list pricing fell **20% for input** and **one-third for output**. Vercel's 50% AI Gateway discount applies to the lower rates through **September 18** across Default, Flex, and Priority tiers. Existing requests keep the same model ID and receive the new rate automatically. For agent workloads up to **272K tokens**, recheck cost assumptions for long investigations and high-output coding runs. The listed rates exclude special pricing details for longer contexts, caching, cache writes, and US regional use, though Vercel says those move proportionally. BYOK traffic is billed under the builder's OpenAI rate.
This revises the economics of the earlier GPT-5.6 Sol promotion: the same temporary 50% gateway discount now compounds with lower list rates, so existing evaluations and production runs become cheaper without migration. It strengthens the case for rerunning cost estimates, especially for long agent sessions, while leaving the prior caveats about BYOK, special context pricing, and the September 18 deadline intact.