Poolside Launches Laguna S 2.1 Open-Weight Coding Model With 118B Parameters for NVIDIA DGX Spark
Poolside released Laguna S 2.1, an 118B-parameter mixture-of-experts model optimized for agentic coding and long-horizon tasks. The model activates only 8B parameters per token and supports context windows up to 1M tokens, enabling complex planning and self-correction over multi-hour sessions.
Poolside reported benchmark scores of 78.5% on SWE-bench Multilingual and 59.4% on SWE-Bench Pro, outperforming models 4x to 25x larger. Weights are licensed under OpenMDW-1.1 on Hugging Face and accessible via OpenRouter and the company API, with the caveat that optimized NVFP4 quantization runs on a single NVIDIA DGX Spark. The company stated it completed development in 52 days using its internal pipeline.
From the sources (25 posts)
@eisokantToday we are releasing Laguna S 2.1. At 118B total parameters, with 8B active per token, it does the work of models several times its size on agentic coding. It is remarkably persistent across long-horizon tasks. And it is small enough to
@thom_wolfRT @poolsideai: Today we're releasing Laguna S 2.1, our most capable model to date. It's a 118B total parameter Mixture-of-Experts model w…
@willccbbRT @eisokant: Today we are releasing Laguna S 2.1. At 118B total parameters, with 8B active per token, it does the work of models several…
@huggingfaceRT @poolsideai: Today we're releasing Laguna S 2.1, our most capable model to date. It's a 118B total parameter Mixture-of-Experts model w…
@nousresearchThe new Laguna S 2.1 model by @poolsideai is now free for 2 weeks on Nous Portal. At 118B total parameters with 8B active, it's quick to run and the most capable model they've released so far. Try Portal today at
@nielsroggeCongrats @poolsideai on the release of Laguan S 2.1! It sets a new SOTA on SWE-Bench Pro for models with < 128B parameters, and pushes the Pareto frontier on benchmarks such as DeepSWE and Terminal Bench 2.1
@thdxrRT @opencode: Laguna S 2.1 is now free on OpenCode 1M Context · fully open source Poolside's most capable model to date
@nvidiaaiCongrats to the @poolsideai team on their latest release, Laguna S 2.1. It’s open-weight, delivers way beyond its size, runs great locally and you can customize it with NVIDIA NeMo. Go try it out on @OpenRouter or download from @huggingf
@mtslivePoolside just launched Laguna S 2.1, an open-weight coding model that runs on a single NVIDIA DGX Spark. We talked to @eisokant, co-founder and co-CEO of @poolsideai, about why he believes you should be able to own intelligence no one can
@mtsliveSITUATION DETECTED: Poolside has released Laguna S 2.1, a 118B parameter Mixture-of-Experts coding model with 8B active per token and a 1M token context window. Weights are open on Hugging Face under OpenMDW-1.1. Poolside says it runs on a
@gavinsbakerOP
@onusozRT @ben_burtenshaw: new local coding agent model dropped. poolside released laguna s 2.1: 118B MoE, 8B active, 1M context, open weights, op…
@lmsysorg🎉 Day-0 support for Laguna S 2.1 from @poolsideai is now live on SGLang! 118B total params, 1M context, with thinking & no-thinking modes. ✅ Long-horizon persistence: keeps planning, testing, and self-correcting up to ~24h runs with little
@eisokant🧵 below from @PengmingWang with a few observations from building Laguna S2.1. The most interesting one is that Laguna S2.1 turned out to be very persistent! This means it has some incredible capabilities and also some rough edges, I am re
@techmemePoolside launches Laguna S 2.1, an 118B open-weight model built for agentic coding and long-horizon work, that it says competes with larger open models (@_iainmartin / Forbes) (Visit Techmeme dot com for the link and full context!)
@mervenoyannRT @poolsideai: Laguna S 2.1 is, as far as we can measure, the most capable agentic coding model in its weight class. On Terminal-Bench 2.…
@tekniumPoolside's latest model, Laguna S 2.1 is now available for free for 2 weeks on Nous Portal! Check it out
@eliebakouchvery impressive model for the size by poolside, but imo what's even more impressive is the iteration speed 3 (open) models in 3 months 😮
@aymericroucherLol the madlads at Poolside are showing how release benchmarks should be done: - comparing only to your previous models ⇒ ngmi - comparing to the best similar-class models ⇒ meh - comparing your 118B release to 1T+ flagships AND BEATING THE
@basetenThe American open-weight ecosystem is growing. We're thrilled to support Poolside as they push open-weight coding models forward. Poolside's flagship model family is Laguna: open-weight agentic coding models built for long-horizon tasks. T
@_lewtunRT @poolsideai: Today we're releasing Laguna S 2.1, our most capable model to date. It's a 118B total parameter Mixture-of-Experts model w…
@mervenoyannRT @eisokant: Today we are releasing Laguna S 2.1. At 118B total parameters, with 8B active per token, it does the work of models several…
@clementdelangueRT @jasoncwarner: Today we’re releasing Poolside Laguna S 2.1 It is a 118B-total, 8B-active open-weight model built for agentic coding and…
@clineLaguna S 2.1 is a breakthrough in small model performance, beating models 3x its size. At only 118b param, it beats DeepSeek v4 Pro, Gemini 3.6 Flash, and Thinking Machines Inkling on benchmarks. Try in Cline with model id: poolside/lagun
@tomlikesrobotsRT @eisokant: Today we are releasing Laguna S 2.1. At 118B total parameters, with 8B active per token, it does the work of models several…