Which GPUs and Macs run Laguna S 2.1?

Key facts

Of the setups on this page, only the 2 two-card PCs run it, and only slowly: Laguna S 2.1 needs about 97 GB of graphics memory, so part of it has to run from normal memory. A single card does not have enough graphics memory on its own. In a PC with enough normal memory it also runs slowly on one card.

About 97.3 GB at Q4_K_M, our estimate. Checked . poolside Q4_K_M GGUF.

Try Laguna S 2.1 free in your browser, no login, no install

The cheapest machine on our list that runs it: Mac Studio M5 Ultra 256GB, about $11,565€12,500.

We never pass off an estimate as a benchmark. Commissions never change which hardware we recommend. How we calculate

We don't give a speed for this model yet. This model only uses part of itself for each word (a mixture of experts), and our speed estimate does not handle that yet.

LIVE IN BROWSER · $0 · NO LOGIN

Test Laguna S 2.1 free in your browser before downloading 96 GB

Open full 8-model playground ↗

Click a starter prompt below or type your own question to watch Laguna S 2.1 stream side by side against Gemma 4 31B right on this page:

Get a 1-line alert when a faster quant for Laguna S 2.1 drops — or when GPUs that run it drop >10%

Get a 1-line alert when a new low-bit GGUF/MLX quant fits 128–256 GB rigs or when multi-GPU & Mac Studio prices drop >10% — weekly, zero spam.

Which graphics cards run it

Seven popular graphics cards and two PCs with two cards each, at the standard quality (Q4). Click one for the full answer: memory, speed and price.

Runs slowly2

Part of the model spills into normal memory, which is much slower.

Doesn't fit8

Not enough graphics memory: it needs about 97.3 GB. In a PC with enough normal memory it runs slowly: PC with RTX 5090 32GB, about $7,834€8,467.

Every model on every card

Which Macs and mini PCs run it

Macs and mini PCs from the picker, at the standard quality (Q4). A Mac shares its memory with the rest of the computer. Each name opens that machine's page.

Runs well1

The whole model fits in the memory the graphics can use.

Cheapest hardware for each size

From all 43 graphics cards, PCs and Macs on our list. "Runs slowly" means part of the model runs from normal memory: cheaper, but much slower. A smaller download needs less memory and answers a little less well.

Prices excl. shipping unless marked. Euro prices incl. VAT where the seller charges it. Dollar prices are converted from our euro prices without the 21% EU VAT, before US sales tax.

Prices moved: see Building a local LLM rig in the RAMageddon.

About Laguna S 2.1

118B total, 8B active. Parameter count is an estimate. More parameters usually means a smarter model that needs more memory.

Licence:
OpenMDW-1.1 source
Longest conversation:
about 1573 pages of text (1,048,576 tokens)
Input:
text
Sources and method

Poolside's coding mixture of experts: 118B total, about 8B active, 256 routed experts with 10 per token plus one shared expert, context 1,048,576. Open weights under OpenMDW-1.1. 12 of 48 layers attend to the whole context; the other 36 use a 512-token window. The DFlash draft (2.23 GB) is not in these file sizes.

Ran Laguna S 2.1 on your own hardware? Send us your llama-bench result.

Parameter source · Architecture source · Model page

Memory at each quality level

Each download of Laguna S 2.1 is a different size. The total adds the file, the memory a conversation of about 12 pages of text (8,192 tokens) takes, and 0.8 GB of working space. A measured file size is labelled; the rest is our estimate. The last column is the cheapest machine that holds the total.

Swipe the table sideways to see every column, up to the cheapest machine.

Memory per quant for Laguna S 2.1
Quality (file)File sizeConversationWorking spaceTotalCheapest machine that holds it
F16235.20 GBmeasured GGUF0.5 GB0.8 GB236.5 GBNo machine on our list runs this size.
Q8_0128.75 GBmeasured GGUF0.5 GB0.8 GB130.0 GBMac Studio M5 Ultra 256GB, $11,565€12,500, price unverified
UD-Q8_K_XL128.11 GBmeasured GGUF0.5 GB0.8 GB129.4 GBMac Studio M5 Ultra 256GB, $11,565€12,500, price unverified
UD-Q6_K_XL107.08 GBmeasured GGUF0.5 GB0.8 GB108.4 GBMac Studio M5 Ultra 256GB, $11,565€12,500, price unverified
Q4_K_M96.03 GBmeasured GGUF0.5 GB0.8 GB97.3 GBMac Studio M5 Ultra 256GB, $11,565€12,500, price unverified
UD-Q5_K_M87.86 GBmeasured GGUF0.5 GB0.8 GB89.1 GBMac Studio M4 Max 128GB, $3,683€3,290 (used), price unverified
UD-Q4_K_XL73.40 GBmeasured GGUF0.5 GB0.8 GB74.7 GBMac Studio M4 Max 128GB, $3,683€3,290 (used), price unverified
UD-Q3_K_XL54.09 GBmeasured GGUF0.5 GB0.8 GB55.4 GBMac Studio M4 Max 128GB, $3,683€3,290 (used), price unverified
UD-IQ2_M37.27 GBmeasured GGUF0.5 GB0.8 GB38.5 GBMac mini M4 Pro 64GB, $2,116€1,890 (used), price unverified
UD-IQ1_S33.77 GBmeasured GGUF0.5 GB0.8 GB35.0 GBMac mini M4 Pro 64GB, $2,116€1,890 (used), price unverified

speed: not estimated for MoE yet.

Speeds and memory are our estimate from the hardware specs, not a measured test. Prices are dated and may change. How we calculate

Questions

Can I test Laguna S 2.1 online for free in my browser without downloading 96 GB?
Yes. Laguna S 2.1 (118B total, 8B active, parameter count is an estimate) is one of the 10 free open models at https://localllmrig.com/ and in the inline test box on https://localllmrig.com/can-i-run/laguna-s-2.1. You can send it prompts side by side against Recursive Flint (https://localllmrig.com/can-i-run/flint) or Gemma 4 31B (https://localllmrig.com/can-i-run/gemma-4-31b) using none of your own graphics memory.
Can I run Laguna S 2.1 on 24GB?
Not in full. Laguna S 2.1 (118B total, 8B active, parameter count is an estimate) needs about 97.3 GB at Q4_K_M (96.03 GB measured GGUF + 0.48 GB conversation memory at 8,192 tokens + 0.80 GB working space = 97.31 GB / 97.3 GB total estimate), and a 24 GB card leaves about 23.4 GB usable. The smallest listed file is UD-IQ1_S (33.77 GB measured GGUF, 35.0 GB total estimate), still above the 23.4 GB usable on a 24 GB card.
Can I run Laguna S 2.1 on a 128GB Mac or 64GB Mac mini?
At Q4_K_M (96.03 GB measured GGUF, 97.3 GB total estimate), it is 1.3 GB over the 96.0 GB usable on Mac Studio M4 Max 128GB and Mac Studio M5 Max 128GB. Q4_K_M fits Mac Studio M5 Ultra 256GB (192.0 GB usable, €12,500 / $11,565, price unverified). These measured files fit a 128 GB Mac Studio (96.0 GB usable; the cheaper 128 GB Mac is Mac Studio M4 Max 128GB at €3,290 / $3,683, price unverified): UD-Q5_K_M (87.86 GB measured GGUF, 89.1 GB total estimate), UD-Q4_K_XL (73.4 GB measured GGUF, 74.7 GB total estimate), UD-Q3_K_XL (54.09 GB measured GGUF, 55.4 GB total estimate). On Mac mini M4 Pro 64GB (48.0 GB usable, €1,890 / $2,116, price unverified), UD-IQ2_M (37.27 GB measured GGUF, 38.5 GB total estimate) and UD-IQ1_S (33.77 GB measured GGUF, 35.0 GB total estimate) fit.
Can I run Laguna S 2.1 on a Dual RTX 3090 or Dual RTX 4090 (48 GB VRAM)?
At Q4_K_M (97.3 GB total estimate), Dual RTX 3090, 64 GB RAM and Dual RTX 4090, 64 GB RAM each leave 46.8 GB of graphics memory usable, so the standard file does not fit in graphics memory and would run by offloading to system RAM. These measured files fit inside that 46.8 GB without offload: UD-IQ2_M (37.27 GB measured GGUF, 38.5 GB total estimate), UD-IQ1_S (33.77 GB measured GGUF, 35.0 GB total estimate). Dual RTX 3090, 64 GB RAM is €3,592 / $3,808, price unverified. Dual RTX 4090, 64 GB RAM is €5,470 / $5,910, price unverified.
How fast is Laguna S 2.1 on an RTX 5090?
We don't give a speed for this model yet. This model only uses part of itself for each word (8B active of 118B total, a mixture of experts), and our speed estimate does not handle that yet.
What is Laguna S 2.1's architecture, context length, and license?
Laguna S 2.1 is a 118B-total, 8B-active mixture-of-experts model with a maximum context of 1,048,576 tokens (about 1,573 pages) under OpenMDW-1.1. The model notes describe 256 routed experts with 10 per token plus one shared expert. 12 of its 48 layers attend to the full context and 36 layers use a 512-token sliding window, so the 8,192-token conversation cache is about 0.48 GB (0.5 GB rounded, our estimate). The model notes say the DFlash draft (2.23 GB) is not included in these file sizes.
Can I try Laguna S 2.1 online for free?
Yes. Laguna S 2.1 is on the free chat on the home page. Pick it there, ask a question, and read the answer in your browser. No login and no install.
Do I need a GPU to test it?
No. The free chat runs in your browser. It does not download the model and it does not use a graphics card on your computer. A graphics card only matters if you run the model yourself.
Do I need an account to try Laguna S 2.1?
No. The free chat needs no account and no login.