AIAny
AI Model2026
Icon for item

Laguna XS 2.1

A Mixture-of-Experts causal LLM (33B total, 3B active) tuned for agentic coding and long-horizon workflows; offers 262K-token context, mixed sliding-window/global attention, FP8 KV-cache and native preserved 'thinking' for tool-assisted agents, with local-ready quantized checkpoints.

Introduction

Most coding and agent workflows hit two hard limits: context length for long-horizon state and latency/memory for local inference. Laguna XS 2.1 targets that gap by combining a compact MoE design with a very large context window and engineering optimizations (FP8 KV cache, mixed SWA/global attention) so you can run agentic coding agents locally without sacrificing multi-step tool reasoning.

Key Capabilities
  • High-context agentic reasoning: preserves explicit "thinking" blocks and supports interleaved reasoning between tool calls, so multi-step plans and tool interactions remain coherent across long sessions.
  • MoE with low activation footprint: 33B total parameters but ~3B activated per token via 256 experts, enabling strong capability with reduced runtime memory compared to dense models — useful for desktop or small-server inference.
  • Long context + efficiency features: 262k token context, sliding-window/global attention mix, and FP8 KV-cache reduce memory and enable long-horizon code and terminal-style tasks.
  • Local-ready ecosystem support: official support and quantized variants for vLLM, Transformers, TRT-LLM, Ollama and llama.cpp, making practical local deployment and tool integration feasible.
Who it's for and tradeoffs

Great fit if you need an LLM to run locally for agentic coding, multi-step terminal automation, or long-context reasoning where preserving internal "thinking" improves outcomes. It trades raw single-turn SOTA for a balance of multi-step capability, local resource efficiency, and tool-call fidelity. Look elsewhere if you need the absolute top leaderboard single-shot accuracy or prefer fully closed-source commercial models with managed hosting and SLA-backed inference.

Information

Categories

More Items

Hugging Face
AI Model2026

Transcribes speech audio into text across 45 languages, with automatic language detection. Its 0.5B parameters make it a compact option for multilingual transcription using Transformers or vLLM.

Hugging Face
AI Model2026

Combines Korean-English language and image understanding with screenshot-driven browser control in one model. It prioritizes retaining general LLM and VLM capabilities rather than optimizing solely for web tasks.

Hugging Face
AI Model2026

Turns photos into detailed foreground alpha mattes, with particular attention to hair, fur, shadows, and complex edges. Built on a generative image-editing model and distilled to four steps; self-hosted weights are non-commercial only and require substantial GPU memory.