if you have a 128 GB AMD Strix Halo running Windows, you can run Qwen3.8-Flash-Next on my custom 102 GB quant (Q4_K_XL) at ~22 tok/s using MTP speculative decoding. don't abuse browser tabs though... the RAM pool gets razor thin (~1 GB free). enjoy: huggingface.co/bitlamas/Qwen…
Laguna S 2.1 is one of the best models in its class. Consumer platforms like Strix Halo with 128 GB unified DDR5 are often overlooked in LLM benchmarks, so I'm running my own benchmark gauntlets based on real workloads. So far, it's a very promising contender. Full results soon.
Today 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 run on a single NVIDIA DGX Spark.
It is far more capable than anything we have created before, and I think it redefines what a model in its weight class can do.
Laguna S 2.1 is an important model for Poolside. What it represents is even more important.
If, five years ago, I had read a book that said that by 2030 everything economically valuable, scientifically interesting, and personally meaningful would be built on intelligence contracted from three or four companies, I would have called it dystopian science fiction.
We are at a fork in the road of what kind of world we can have.
I believe intelligence should and will become a commodity. The question is whether that intelligence comes from three companies, or from many people who can build it, own it, and shape it.
The open ecosystem will not win by being the best in its own category. No one cares who is king of the open-source kingdom. People want the best intelligence for the task they are trying to do, with the right balance of quality, speed, cost, and control.
If we want a different future, open models have to be on par with, or better than, their closed equivalents.
Laguna S 2.1 is a meaningful step in that direction: capable enough to compete far above its weight class, efficient enough to run on hardware you can own, and open-weight so anyone can build on it.
Open-weighting our models is the contribution we can make today toward a world where intelligence can be built and owned by many. And we will keep doing it.
I am very proud of this team’s work. A big shout out to everyone at Poolside who made this possible, from infrastructure and data to architecture, pretraining, post-training, evaluations, and inference.
Laguna S 2.1 is available today under the OpenMDW-1.1 license, with weights on Hugging Face and access through OpenRouter and our API.
We are building toward a future where the most capable intelligence in the world can be owned and shaped by anyone. Laguna S 2.1 is one step. We are going to keep building until that future exists.
poolside.ai/blog/introducing…
it reads directly from the OAuth endpoint that your Claude Code CLI uses, so you don't need to share any credentials or anything, really. get it: github.com/bitlamas/cc-usage…
90s middle-class kids from non-prosperous countries had an organic path to computer wizardry. i remember at 10 i was editing the Windows registry, running NO$GMB to play Pokémon Red, cracking pirated games.
buying originals was out of reach but at least i ended up learning a lot.






