Poolside

2026Poolside

Laguna-XS-2.1 v2.1

Open Source
mixture-of-experts
coding-agent

Laguna-XS-2.1 is a 33B parameter Mixture-of-Experts (MoE) model designed for agentic coding and complex, long-horizon tasks. Featuring 3B activated parameters per token, it provides enhanced performance on benchmarks like SWE-bench Multilingual while remaining efficient enough to run on local hardware.

33B total parameters with 3B activated per token
Mixed sliding window and global attention layout (3:1 ratio)
Native reasoning support with interleaved thinking capabilities
PricingOpen Source
Version2.1
2026Poolside

Laguna XS 2.1 v2.1

Open Source
Featured
LLM
Coding Agent

Laguna XS 2.1 is a 33B total parameter Mixture-of-Experts (MoE) coding agent model with 3B activated parameters per token, specifically engineered for high-performance agentic coding and long-horizon tasks. It features an efficient architecture with mixed sliding-window and global attention, native interleaved reasoning, and support for a 256K context window, making it highly capable for local execution and complex software engineering workflows.

33B total parameter MoE model with 3B activated parameters
256K-token context window with native interleaved reasoning support
Mixed sliding-window (512) and global attention (3:1 ratio)
PricingFree and open-weight model
Version2.1
2026Poolside

pool v1.0

Open Source
AI Coding Agent
ACP

pool is a versatile coding agent designed by Poolside that functions as a standalone terminal application or integrates with any Agent Client Protocol (ACP) compatible editor. It supports various operational modes, including interactive terminal sessions, non-interactive CI/CD execution, and acts as both an ACP client and server to drive complex software development workflows.

Terminal-based interactive coding agent with support for slash commands, file fuzzy search, and shell mode
ACP-compatible design for seamless integration with IDEs like Zed, JetBrains, and Xcode
Non-interactive execution mode (pool exec) for CI/CD pipelines and headless automation
Version1.0
RegionUnited States
2026Poolside

Laguna S 2.1 v2.1

Open Source
AI Agent
Coding Assistant

Laguna S 2.1 is an advanced 118B Mixture-of-Experts (MoE) agentic coding model featuring 8B activated parameters per token. It is specifically designed for long-horizon software engineering tasks and complex reasoning, capable of operating within a 1M token context window.

Mixture-of-Experts (MoE) architecture with 118B total parameters and 8B activated parameters
Supports a 1M token context window in both thinking and no-thinking modes
Agentic capabilities optimized for long-horizon coding and complex software engineering
Version2.1
RegionUnited States

🚀 Join the AI dev community — follow us everywhere

© 2026 MARKTECHPOST AI MEDIA INC. All rights reserved.Terms & ConditionsPrivacy Policy
Beta Mode