TAAFT
Free mode
100% free
Freemium
Free Trial
Prompts Deals
Companies
poolside
Models
Laguna S 2.1

Laguna S 2.1

By poolside
Laguna S 2.1 is an open weight Mixture of Experts model with 118B total parameters and roughly 8B activated per token, designed for agentic coding and long horizon work. It uses 48 layers in a 1 to 3 global to sliding window attention ratio, grouped query attention, and a token choice router with 256 routed experts plus one shared expert. The model supports a 1,048,576 token context window and native reasoning, interleaving thinking between tool calls with per request control. It performs strongly on coding benchmarks including Terminal Bench, SWE bench Multilingual, and SWE Bench Pro. Quantized variants (FP8, NVFP4, INT4, GGUF) and a trained speculative decoding draft model are available for lower latency serving. It is licensed under OpenMDW 1.1, permitting free commercial and non commercial use and modification.
New Text Visit model
Released: July 21, 2026

Overview

Laguna S 2.1 is a 118B parameter Mixture of Experts model with 8B activated parameters per token, built for agentic coding and long horizon software engineering work. It has a 1,048,576 token context window, native interleaved reasoning, and mixed global/sliding window attention across 48 layers.

💻Coding✉️Codebase Q&A🐞Debugging🔬Code testing🤖Agents🔧Code refactoring

About poolside

poolside is a foundation-model company building AI for software engineering. It develops proprietary frontier models (Malibu, Point, and the open-weight Laguna XS.2) plus a full agentic platform — IDE/TUI integrations, multi-agent systems, and connectors — deployed on-prem, in customer VPCs, or in air-gapped environments. Targets large enterprises and the U.S. defense industrial base. Mission stated as building AGI via software engineering as the strategic beachhead.

Industry: Artificial Intelligence
Company Size: 322
Location: San Francisco, California, US
Website: poolside.ai
View Company Profile

Related Models

  • Apertus 8B 2509
    Apertus 8B 2509 is a fully open 8B parameter language model by Swiss AI Initiative, pretrained on 15T tokens. It natively supports over 1800 languages including all Swiss national languages, uses only compliant open training data, and achieves performance comparable to closed-source models. Supports long context up to 65K tokens and tool use.
    📝Text generation🌍Translations‍💻Code generation
    Text
    Released 10mo ago
  • SERA 32B
    SERA-32B is Ai2’s flagship 32B Open Coding Agent, a Qwen 3-32B based repository-level coding model trained with Soft Verified Generation that reaches about 49.5% on SWE-bench Verified, rivaling much larger proprietary systems.
    💻Coding📚Code explanations🔍Code reviews💡Coding help
    Text
    Released 5mo ago
  • Voyage 4
    Text embedding model optimized for general-purpose and multilingual retrieval quality. Supports a 32,000-token context window and configurable output dimensions (256, 512, 1024, 2048) via Matryoshka Representation Learning. Part of the Voyage 4 series, which shares a common embedding space enabling asymmetric retrieval across models in the family.
    🔗Embeds🔍Information retrieval🔎Semantic search
    Text
    Released 6mo ago
Last updated: July 22, 2026
0 AIs selected
Clear selection
#
Name
Task