Laguna S 2.1 Now Available on AI Gateway

Poolside's Laguna S 2.1 model has launched on AI Gateway. Users can choose between a free version (poolside/laguna-s-2.1-free) and a paid version (poolside/laguna-s-2.1) that offers higher throughput and rate limits.

Laguna S 2.1 is an open-weight Mixture-of-Experts (MoE) model supporting a context window of up to 1M tokens, and it supports both 'Thinking' mode and 'No-thinking' mode. It is especially specialized for Agentic Coding and long-running tasks.

Key features and performance are as follows:

• Specialized areas: Writing and debugging code, running tests, building browser-based tooling, MLOps pipelines, and AI research
• Benchmark performance (Thinking mode): Terminal-Bench 2.1 70.2%, SWE-bench Multilingual 78.5%, SWE-Bench Pro 59.4%

AI Gateway provides a unified API for model calls, including usage and cost tracking, retries, failover, and performance optimization. It also offers features such as custom reporting, Zero Data Retention support, API key budget settings, and routing rules, while passing through provider pricing as-is with no separate platform fee.