Laguna XS.2
#10 in Lütte & Edge-Modellepoolside · siet 2026-04-28 · 16× · tolest 30. Juni 2026
24
Momentum
Laguna XS.2 is a Mixture-of-Experts language model by poolside with 33 billion total and 3 billion active parameters per token, built entirely from scratch for agentic coding and long-horizon tasks on local hardware. It is poolside's first open-weight release, trained through pre-training, post-training, and reinforcement learning stages. The architecture combines sliding-window attention in 30 out of 40 layers with 10 global attention layers (3:1 ratio) and an FP8-quantized KV cache. The model runs on a single consumer GPU or a Mac with 36 GB of RAM and natively supports interleaved reasoning between tool calls.
Momentum-Verloop
04.04.03.07.
Features
| Key Benchmark (%) | 68.2% on SWE-bench Verified; 44.5% on SWE-Bench Pro; 30.1% on Terminal-Bench 2.0 |
| Context Window (Tokens) | 256,000 tokens (262,144) |
| License | Apache 2.0 (open weights, free for commercial & non-commercial use) |
| Multimodality | Not multimodal – pure text-generation model for agentic coding |
| Platform | Hugging Face, Ollama (MLX), vLLM, SGLang, TRT-LLM, OpenRouter, Poolside API, Zed/JetBrains via ACP |
| Price per 1M Tokens | $0.10 input / $0.20 output (standard on OpenRouter); temporarily free ($0/$0) via Poolside API & OpenRouter |
| Release Date | April 28, 2026 |