Models / poolside/ Laguna S 2.1

Laguna S 2.1

poolside · released Jul 13, 2026 · poolside/Laguna-S-2.1

Input: text. Output: text.InputOutput
Type
Open weightsopenmdw-1.1
Params
118B
Context
1M

about 786K words of context · download allowed, licence restricts use

Our take

Written Aug 2, 2026

Laguna S 2.1 is a 118-billion-parameter text model from poolside with a one-million-token request limit and a restricted open licence. It is built for very long documents, though no quality scores have been published yet.

Who should pick it

Choose this when you need to process text documents of one million tokens or more, and when the OpenMDW 1.1 licence terms fit your use case. Pick the faster poolside tier if latency matters more than a modest output premium. Skip it if you need measured quality data, multimodal input, or a permissive licence like Apache or MIT.

The case for it

  • One-million-token request limit — the only figure we hold for this model, and extremely long among models with disclosed length.
  • Three hosted offers with competitive relative pricing in its class, including a faster tier at 15 tokens per second.

The case against it

  • No benchmark scores in our data, so there is no measured quality or speed data to validate performance claims.
  • The slower tier at 6 tokens per second costs nearly as much as the 15 tokens per second tier, making it poor value.
  • Only three tracked offers, with throughput unverified on one of them.
00

How good is it?

We hold no score for this model.

We look for every model we track on every board we watch, and none of them has turned up Laguna S 2.1 — so there is no intelligence, coding, agentic or writing score to show you, not a low one, none.

The boards we watch →

01

Can you run it yourself?

Fits in memory
weights load entirely on the card
Spills to system RAM
some weights offload; much slower
Too large
will not load even with offload
est
size is calculated; the verdict could change by 10%
A card many people ownToo large

GeForce RTX 4090 · 24 GB

Weights at Q4_K_M96 / 24 GBmeasured
Usable contextNot calculated for spilled setups
Decode speedNot estimated for spilled setups

Too large for this card. The weights do not fit even with part of them offloaded to system memory.

One step upToo large

Apple M1 Pro (16-core GPU) · 32 GB

Weights at Q4_K_M96 / 32 GBmeasured
Usable contextNot calculated for spilled setups
Decode speedNot estimated for spilled setups

Too large for this card. The weights do not fit even with part of them offloaded to system memory.

Comfortable fit

On a MacFits in memory

Apple M2 Ultra (76-core GPU) · 192 GB

Weights at Q4_K_M96 / 192 GBmeasured
Spare memory43.5 GB spare
Usable context131K of 1M
Decode speed7 tok/sest

Room to spare. 43.5 GB spare means a 10% error in the size would not change the answer.

Your hardware
Checking your profile…

Memory use by level

Against a 24 GB card.

F16
235.2 GBmeasured
Too large
Q4_K_M
recommended
96 GBmeasured
Too large
Q5_K_M
87 GBest
Too large
Q8_0
127.7 GBmeasured
Too large

Check against your own machine → · All 71 devices, with every size →

02

Or rent it from someone else

Cheapest published offer

Cheapest of 3 live listings. Picked at the widest standard context we hold, within one quantisation slice, so the numbers beside it are a price one host actually charges.

per 1M tokens
$0.090 in / $0.18 out
Context served
1M
Throughput
Not measured
Current provider offers with price, context and prompt-privacy answers
ProviderIn / out per 1M tokensContextThroughputTrains on promptsLogs promptsZero retention
OpenRouter$0.090 / $0.181Mnot measuredUnknownUnknownUnknown
Poolsidefp4$0.090 / $0.181M14 tok/sNoYesunknown periodUnknown
Poolsidebf16$0.10 / $0.201M15 tok/sNoYesunknown periodUnknown

Across the 3 listings we hold: 2 say they do not train on prompts, 0 say they do and 1 do not say. 0 appear in the zero-retention registry we check; the rest are unknown to us rather than confirmed either way.

What each host's API supports

From the parameter list each endpoint publishes. Streaming is omitted: nothing we hold reports it, for any model.

Supported
Not supported
Not published
host gave no parameter list
API features per host
ProviderTool callingJSON outputStrict schema
OpenRouter
Poolsidefp4
Poolsidebf16

Tool calling: 3 of 3 listings say yes. JSON output: 0 of 3 listings say yes, 3 say no. Strict schema: 0 of 3 listings say yes, 3 say no.

03

When we formed this view

Dates behind this page

Jul 30, 2026Price changeOpenRouter cut Laguna S 2.1 pricing by 10%input −10% ($0.10 → $0.090 per 1M tokens); output −10% ($0.20 → $0.18 per 1M tokens); cache read −10% ($0.010 → $0.009 per 1M tokens)
Jul 26, 2026ListedListed on LLMapfirst indexed by our pipeline
Jul 21, 2026ReleaseLaguna S 2.1 listedNew model detected on OpenRouter: poolside/laguna-s-2.1
Jul 13, 2026AnnouncedLaguna S 2.1 announced by poolside

Prices last checked 35h ago

What we do not know about this model yet

  • No board we watch has turned up a score, so we hold no quality figures at all.
  • Nothing we hold says whether an endpoint streams, so we do not show it either way.
  • 1 of 3 listings do not say whether they train on prompts.
04

Licence and identifiers

What the licence allowsopenmdw-1.1, what it allows commercially, and the identifiers you need to pull this model — its Hugging Face repo, our slug and a machine-readable card.

Licence

openmdw-1.1

restricted_openCustom licence — review the terms

License tag "openmdw-1.1" imported from Hugging Face; terms pending curation.

Identifiers

Architecture
Mixture of experts
Modality record
text->text
Catalogue slug
poolside-laguna-s-2-1

Machine-readable model card (omc.json) →

Something wrong on this page? Tell us