Artificial intelligence news.
Straight from the original announcements.

Vercel Agent sessions now cost 67% less on average, company says

October 27, 2026Source: Vercel
Vercel Agent sessions now cost 67% less on average, company says
Image: Vercel

Vercel says Agent sessions in the dashboard and Slack now cost 67% less on average. The company rebuilt how Agent manages context, which reduces the tokens needed per session and improves prompt-cache reuse.

Key points
  • Vercel says Agent sessions in the dashboard and Slack now cost 67% less on average.
  • The savings come from reduced token use per session and better prompt-cache reuse.
  • The improvements are live and require no action from users.
  • Billing rates are unchanged: model provider inference rates plus a Vercel Agent fee of $0.25 per million tokens.
  • Usage and costs can be viewed in the Agent section of the dashboard.

What Vercel changed

According to Vercel, the savings come from a rebuild of how Agent handles context. The changes are already live, and users do not need to take any action to benefit from them.

The company lists four areas of improvement:

  • Workflow-based CLI skills: Updated Vercel CLI skills give agents clearer guidance for common workflows and error recovery. Vercel says other agents that use these skills also benefit.
  • Incremental context management: Agent manages context more efficiently as conversations grow, including long Slack threads.
  • Improved prompt caching: Agent preserves more reusable context across turns, so more of it can be read from the cache.
  • More efficient coding and planning: Updated guidance helps Agent plan changes and write code with fewer tokens.

Billing rates stay the same

The lower cost does not come from a price cut. Vercel states that Vercel Agent's billing rates are unchanged. Paid sessions are billed at the model provider's inference rates plus a Vercel Agent fee of $0.25 per million tokens.

Because charges are tied to token usage, a session that needs fewer tokens costs less under the same rates. Better cache reuse also means more of the context can be read from the cache across turns rather than being processed again. The 67% figure is an average reported by Vercel, so the savings in an individual session may differ.

What it means for users

Teams that run Agent sessions in the Vercel dashboard or in Slack should see lower costs without changing their setup. Long Slack threads are specifically named as a case where context is now handled more efficiently.

Users can check their own numbers in the Agent section of the dashboard, where Vercel says usage and costs are shown. The updated CLI skills may also be useful beyond Vercel's own product, since the company says other agents using them benefit too.

The changelog entry credits Kathryn Middleton, Shilpa Apte and Pranav Kanchi as contributors.

Related news

Vercel Adds DeepSeek V4.1 Flash 1017 Ultrafast to AI Gateway
Products and toolsNovember 7, 2026

Vercel Adds DeepSeek V4.1 Flash 1017 Ultrafast to AI Gateway

Vercel has added DeepSeek V4.1 Flash 1017 Ultrafast to its AI Gateway through Fireworks, giving developers access to a model that can generate upwards of 1,000 tokens…

Vercel adds DeepSeek Search and Fetch to AI Gateway
For developersNovember 6, 2026

Vercel adds DeepSeek Search and Fetch to AI Gateway

Vercel has added DeepSeek Search and DeepSeek Fetch to AI Gateway, allowing developers to give models web search and page retrieval tools through one AI Gateway API key.…

Vercel adds streaming generative UI to Chat SDK agents
For developersNovember 6, 2026

Vercel adds streaming generative UI to Chat SDK agents

Vercel says Chat SDK agents can now stream generative UI with json-render, allowing agents to generate cards from typed JSON specifications. Chat SDK adapters then…