10x more capacity for Laguna S 2.1 on AI Gateway

Poolside’s Laguna S 2.1 now has 10x more capacity on Vercel’s AI Gateway. The increase covers both the paid model id poolside/laguna-s-2.1 and the free model id poolside/laguna-s-2.1-free, so callers can send far more requests against either path without changing providers.

Capacity here is an AI Gateway quota change, not a new model cut. You still target Laguna S 2.1 by setting the model string to poolside/laguna-s-2.1 or poolside/laguna-s-2.1-free. Free and paid share the same 10x headroom, so routing choice stays about access tier rather than whether the higher limit applies.

That matters for builders running high-volume agentic coding and long-running tasks. Those workloads fan out many tool calls and keep sessions open; the prior cap was the bottleneck more often than model quality. A 10x raise on both free and paid ids reduces rate-limit retries and stalled agent loops when request volume spikes.

On the market side, Poolside’s Laguna S 2.1 is now easier to run at scale through Vercel’s AI Gateway as a single integration surface. Teams already on the Gateway get the capacity win by model id alone, which lowers the cost of trying or sticking with Poolside for agent-heavy coding pipelines versus wiring a separate high-throughput path.

Practical next step: point agent and batch jobs at poolside/laguna-s-2.1-free or poolside/laguna-s-2.1 and recheck throughput under your real concurrency. Watch residual Gateway or account limits, error rates under sustained load, and whether free-tier behavior still differs from paid once volume is no longer the first constraint.