Birbla

Join the discussion

Write your take first — we'll ask for email only when you're ready to publish.

Sign in
  • Hacker News
  • This is INSANE. How did they do this?
    by megavon
  • "It went from the start of training to launch in under nine weeks..."

    This is pretty impressive.

    by kamranjon
  • Are all AI labs in Google's weight class in crawling and ranking the Web's content? I know OpenAI has contractor subject matter experts in all topics.
    by aitchnyu
  • "What we've done in this model is not necessarily add more intelligence, but improve the behaviors that lead to a more capable model: more verification, less taking things for granted, not declaring victory early, and being more persistent.”

    +

    https://poolside.ai/assets/laguna/laguna-m1-xs2-technical-re...

    by eisokant
  • I love this. Is it possible to give a feel of how this stacks up to the good old Opus 4.5 in coding quality? For me that was the turning point where agentic coding in Claude Code etc became usable. Have we hit that threshold?
    by river_otter
  • Having played with it for like 3 hours now....I'm probably moving from CC to this
    by megavon
  • Model Looks amazing!

    Even more important, subjectively, is that this model will run very well on Strix Halo (e.g. Framework Desktop), DGX Spark kinds of devices. Looking forward to Unsloth dynamic mtp quants.

    P.S. Looking at the HF release they already offer Q4_K_M and DFlash drafter for speculative decoding!

    by Iolaum
  • I hope all models going forward come with a dflash drafter so we don't have to train one up separately.
    by verdverm
  • Really impressive signal that this 128B model can beat DeepSeek V4 (1.6T) on most coding benchmarks!

    Also, I really like Poolside's habit to compare not only to other top models in its weight class (others don't do it, looking at you Mistral), but also to the very top open-weight models, even much bigger ones like the 2.5T Kimi-K3!

    by aubanel
  • !! Be careful when testing the model.

    A lot of people are testing it, and reporting disappointed results / benchmaxxxing claim. But do not realize that thinking has a issue with the default configuration.

    Important - make sure that THINKING is enabled. By default it wasn't although I was passing the flag --default-chat-template-kwargs '{"enable_thinking": true}' in vllm recipe. The generation_config.json file that is included has by default max_new_tokens as 32k which seems to be cutting off thinking altogether so increase it. At first I was very disappointed with the output I was seeing, but once thinking is enabled, the code quality seems to be MUCH better. More real world testing to be done.

    https://www.reddit.com/r/LocalLLaMA/comments/1v2pg99/laguna_...

    by benjiro29
  • Thanks for posting this, it made a huge difference tweaking the recipe.
    by nshotton
  • Even the official provider on OpenRouter seems to have this issue. Hope it's an easy fix for them.
    by voxgen
  • Looks like the default chat template was updated on HF to enable this by default shortly after you posted this :-)
    by d2p
  • This is exactly the kind of model that's been needed in the middle. Realistically self-hosted, Good Enough intelligence, MoE so it's fast on limited bandwidth systems like Strix Halo and DGX Spark.

    For a while there's been nothing to run on my Strix Halo that's notably better than what I can run on my dual 32GB GPU desktop (Gemma 4 or Qwen 3.6 dense models), but this seems likely to be the step up in size that actually works better than those.

    by SwellJoe
  • On Strix Halo, have a look at the work antirez is doing with dwarfstar (antirez/ds4) for deepseek v4 flash at reasonable speeds and quality.
    by mandelken
  • Whoa whoa whoa, 118b params, 8b active MOE, long context reasoning, open weights - music to my ears. Hadn't heard of this lab before but I am very excited, will definitely try this out tomorrow - this is a real sweet spot I think in terms of model size and performance.
    by kamranjon
  • If the numbers are legitimate then our prayers have been heard
    by svclaws
  • Looks impressive, and this size fits achievable home hardware.

    That said, if someone would kindly quantise this down for the 64GB paupers, that would be appreciated. (I know there’s likely degradation, but some people reported good results with a 2 bit version of Qwen 3.5 122B, and this is starting from a higher point. Would be interesting to try, at least.)

    Edit: someone in the process of doing so: https://huggingface.co/vcruz305/Laguna-S-2.1-GGUF

    by mft_
  • This is interesting: https://github.com/tanishq-dubey/macos-laguna-s2.1

    Someone has benchmarked a wide range of different quants.

    by mft_
  • unsloth is always here to save the day https://huggingface.co/unsloth/Laguna-S-2.1-GGUF
    by nunodonato
  • > if someone would kindly quantise this down for the 64GB paupers

    The Q4_K_M is 75GB. I'm exactly at 64GB, and I wouldn't quantize it further. Instead, do partial weight residency and stream the rest from SSD.

    by Archit3ch
  • They have also published smaller 33B model called Laguna XS 2.1, its Q4 gguf is 20GB.

    https://huggingface.co/poolside/Laguna-XS-2.1-GGUF/tree/main

    by yogeshp
  • The tool I've been using, llm-compressor, can quant models that do not fit in memory (use the sequential pipeline)

    https://github.com/vllm-project/llm-compressor

    my setup to help you on your way: https://github.com/verdverm/quantr

    Though it seems these will not be needed as Poolside has published quants & dflash with their models.

    by verdverm
  • Hey, this model is not a joke! Exciting, we already got a usable PR of work out of it.

    https://github.com/mozilla-ai/otari/pull/348

    by river_otter
  • Incredible. This is definitely the launch of the day. Just crushing Google's releases.

    The pricing here is incredible. This is the first US release that's competitive with DeepSeek V4 Flash. Very excited about this.

    by mchusma
  • Testing it now. At the very least, competitive with DS4-Flash indeed. On my small (and per Sol's words, _very_ semantically dense) C test codebase, it found things that only gpt-5.2 managed to find back in the day, but also made a stupidly incorrect initial observation that a memfd_create()/mmap was used for IPC (funnily enough - sol missed that as well in its review, until I pointed it out). Re: the claims vs deepseek v4 - both flash and pro are expected to get a "general availability" release very soon (i.e. well-"post-trained"), so things can change in a... well, flash, as per usual in the current environment.

    Anyways, keep 'em coming.

    by Lwerewolf
  • What harness/quant did you use for testing?
    by ilc
  • This seems like a great option for local usage. With 16 GiB of VRAM on the official (https://huggingface.co/poolside/Laguna-S-2.1-GGUF) Q4_K_M quant (which seems to be recommended) I can get about 50 t/s in and 10 t/s out. Definitely not winning any speed benchmarks but totally usable for background tasks.

    The intelligent and decision making looks good overall. It seems to roughly compare to unsloth/Qwen3.6-27B-GGUF:UD-Q8_K_XL while being way faster on my hardware. It does often get stuck considering the same decision over and over again but maybe this is how it makes the better decisions.

    However it does seem to have a bad habit of corrupting things (most often my name). It will rewrite kevincox to kevinc or kevcox or kevancox. Then fail to read directories or rewrite code with the wrong string and mess it up. Then it can't identify it's mistake.

    Maybe I'll have try try out the Q8 model, but the description seems to push away from this one.

    by kevincox