Models/Poolside/Laguna S 2.1

Laguna S 2.1

Poolside’s current S-class checkpoint. #12 on OpenRouter This Week at 1.67T (−5% week-over-week, not share) on the free slug. Open OpenMDW 1.1 weights. No official DeepSWE, AA, or vals card.

01

Identity

1,048,576Context in
131,072Context out
2026-07-21 (OpenRouter created 1784652683)Released
OpenWeights
Canonical name
Laguna S 2.1
Aliases
laguna-s-2.1, poolside/laguna-s-2.1
Continuum hosted id
Not listed as a Continuum hosted id
OpenRouter slug
poolside/laguna-s-2.1
Hugging Face
poolside/Laguna-S-2.1
Modalities
text
Open weights Not on Continuum host list Vendor lab

Not listed as a Continuum hosted id. OpenMDW 1.1 on the official repo. OpenRouter catalog created unix 1784652683 (21 Jul 2026). 1,048,576 in / 131,072 out; text→text. Hugging Face id poolside/Laguna-S-2.1.

HF config.json (fetched 19 Aug 2026): LagunaForCausalLM, 48 layers, hidden 3072, 256 experts, 10 experts/token, 1,048,576 positions. Chat template is present. No arXiv tag on the Hub card we fetched.

Week-token rank is #12 on OpenRouter This Week (1.67T, −5% WoW) on the free slug. That is volume, not a coding score.

02

Should I use this for coding agents

Use Laguna when you want Poolside’s official open checkpoint. Do not invent a coding percent. No official DeepSWE, AA, or vals row: those chips are omitted. The only live dollar table we fetched is OpenRouter.

No independent row for this SKU

No independent board lists this exact SKU as of 2026-08-19.

No other card from this lab carries an independent board row either, so there is no sibling worth pointing at.

Boards checked: DeepSWE official (2026-08-13), Artificial Analysis Intelligence Index, vals.ai hero index. Every card that does carry a row is on the coding leaderboard.

DeepSWE

No official DeepSWE mini-swe-agent row for this identity on the public board we fetched.

omitted, not invented as of 2026-08-13 source
Artificial Analysis

No Intelligence Index integer fetched for this identity. Chip omitted.

omitted, not invented as of 2026-08-19 source
Vals Index

We did not open a vals.ai card for this identity. Chip omitted.

omitted, not invented as of 2026-08-19 source

vals 404 as of 19 Aug 2026. Chip omitted.

Named benches, printed only

Copied from AA or vals HTML we opened. 0.0% placeholder bars are omitted. Do not average these into the hero tiles, and do not invent CI, $/task, or steps for Artificial Analysis.

BenchPrintedNoteSourceURLAs of
OpenRouter list pair $0.09 / $0.18 per 1M Catalog row. No official Poolside per-M table extracted. OpenRouter catalog source 2026-08-19
OpenRouter cache read $0.009/M Printed cache-read field on the same catalog row. OpenRouter catalog source 2026-08-19
03

When not to use it

Trust this before you buy

  • No official DeepSWE, AA, or vals row (AA and vals 404). Week-token rank #12 is volume, not a coding score. Do not treat 1.67T as “better than Fable.”
  • OpenMDW 1.1, not MIT. No official GGUF on this repo. Community quants are community.
  • Not a Continuum hosted id. The week-token figure is the free slug’s volume, not a hosted Continuum id.

Plus is $25/mo with $25 weekly hosted usage. The Mac app stays free with your own keys. Get Plus is not Download for Mac.

04

Artifact and Hugging Face downloads

Artifact

SPDX / license
OpenMDW-1.1
Total params
HF safetensors.total: 117,561,977,600 stored tensors
Architecture
LagunaForCausalLM · 48 layers · 256 experts · 10 experts/tok · hidden 3072 · 1,048,576 positions
Native precision
HF tensors BF16
Files
46 safetensor shards (model-00001-of-00046 …)
Repo size
HF API usedStorage 235,129,327,775 bytes (219 GiB)
HF created
2026-07-13T13:58:35Z
Chat template
Jinja present (chat_template.jinja).
Paper
No arXiv tag on the Hugging Face API payload we fetched.
Official HF repo
poolside/Laguna-S-2.1
HF downloads
121,300
HF likes
972

Community quants: Community quants are community.

Official weights

This is the model repo, not Download for Mac. Downloads and likes from the Hugging Face API on 2026-08-19.

Hub file tree

Safetensor shards
46
Shard bytes
235,128,558,560 bytes
Other file bytes
10,713,462 bytes
Tree file count
61
Hub usedStorage
235,129,327,775 bytes

usedStorage and the sum of .safetensors sizes can differ (LFS pointers, non-shard files, duplicate copies). Cited separately. Source: recursive tree API, 2026-08-19.

PathBytes
model-00001-of-00046.safetensors5,079,805,456
model-00002-of-00046.safetensors5,118,428,616
model-00003-of-00046.safetensors5,115,054,984
model-00004-of-00046.safetensors5,115,054,984
model-00005-of-00046.safetensors5,115,054,984
model-00006-of-00046.safetensors5,115,054,984
model-00007-of-00046.safetensors5,115,054,984
model-00008-of-00046.safetensors5,115,054,984
model-00009-of-00046.safetensors5,115,054,984
model-00010-of-00046.safetensors5,115,054,984
model-00011-of-00046.safetensors5,115,055,728
model-00012-of-00046.safetensors5,115,055,736

Showing the first 12 of 46 safetensor shards. The tree also holds 15 non-shard files, none of them listed here. The snapshot was capped at 12 shards when the tree API was read on 2026-08-19, so the remaining 34 are on the Files tab, not on this page. Totals above are the full tree API sum, not this slice. Byte sizes are Hub tree API size fields, never invented.

05

Continuum serving

Not listed as a Continuum hosted id. OpenRouter poolside/laguna-s-2.1.

Continuum hosted id not on the host list

We list first-party and OpenRouter, plus Continuum only when the live public allowlist named the id. This is not a 15-host routing table.

06

Economics

OpenRouter catalog 19 Aug 2026: $0.09 / $0.18 per million, cache read $0.009 /M, 1,048,576 / 131,072, created 1784652683. No DeepSWE $/task.

Price this model Opens the pricing calculator preloaded with Laguna S 2.1.

AA 404 as of 19 Aug 2026. No Index integer, no tok/s, no Index-eval cost.

07

Provenance

ClaimSourceAs of
OR catalog $0.09 / $0.18, cache read $0.009/M, 1,048,576 / 131,072, created 1784652683, hugging_face_id poolside/Laguna-S-2.1OpenRouter catalog2026-08-19
1.67T week tokens (−5% WoW) on the free slugOpenRouter rankings This Week2026-08-19
HF safetensors.total 117,561,977,600; usedStorage 235,129,327,775; 46 shards; likes 972; downloads 121,300; created 2026-07-13T13:58:35Z; OpenMDW-1.1Hugging Face API poolside/Laguna-S-2.12026-08-19
AA / vals 404; DeepSWE board has no exact Laguna S 2.1 rowArtificial Analysis / vals / DeepSWE2026-08-19
OpenRouter id poolside/laguna-s-2.1, context 1,048,576OpenRouter /api/v1/models2026-08-19
HF downloads 121,300Hugging Face API poolside/Laguna-S-2.12026-08-19
46 safetensor shards; shard bytes 235,128,558,560; tree files 61Hugging Face tree API poolside/Laguna-S-2.12026-08-19
Spaces API returned 15; official collection Laguna S 2.1 (4 items, 45 upvotes)Hugging Face Spaces / collections API2026-08-19
08

Compare, FAQ, and Get Plus

No other card in this cluster shares a published DeepSWE official row we can put next to this one. We will not compare on vendor-blog numbers.

Continue through Poolside's model family with Laguna XS 2.1.

FAQ

Why no DeepSWE / AA / vals chip?

The official DeepSWE board, Artificial Analysis, and vals.ai had no fetchable Laguna S 2.1 row on 19 Aug 2026. Omitting is the rule.

Why does the lab blog disagree with DeepSWE or Scale?

Lab posts pick a harness, an effort, a split, and sometimes a private eval set. DeepSWE publishes the official mini-swe-agent row with cost, tokens, and steps. Scale SWE-bench Pro only counts when the public shared-harness board has a row we can fetch. A higher lab number is usually a different test, not a better one.

OpenAI-compatible call

No Continuum hosted id is verified for this model, so this page does not invent a curl target. Use the first-party API or the OpenRouter slug poolside/laguna-s-2.1.

Plus is $25/mo with $25 weekly hosted usage. The Mac app stays free with your own keys. Get Plus is not Download for Mac.