whatmodel.sh / Laguna S 2.1 (118B-A8B)

Laguna S 2.1 (118B-A8B)

+ contribute

Laguna S 2.1 (118B-A8B) is an open-weights language model from Poolside, released July 2026. Its published size is 118B parameters (8B active per token) and a 1M-token context window. Licence: openmdw-1.1.

Catalogue identity notes

Poolside's Laguna S 2.1 (poolside/Laguna-S-2.1 @0f57314, created 2026-07-13, licence openmdw-1.1), between Laguna XS 2.1 (33B-A3B) and Laguna M.1 (225B-A23B) per the card. Hosted access is via third-party gateways only (OpenRouter, Vercel), so api is null.

Model information sources

Release date ↗ · Weights licence ↗

Card: 'a 118B total parameter Mixture-of-Experts model with 8B activated parameters per token' (safetensors metadata 117,561,977,600).

73.4±4.1#57 of 949 LLMsfrom 2 benchmark results across 1 source* thinly tested, treat the rank as provisional

$ how it compares

The LLMs ranked either side of Laguna S 2.1 (118B-A8B), out of 1132 with a score, then more from Poolside.

#	model	score	where	RAM	released
62	Qwen3.7 Max Preview Alibaba	74±3	comparison only
63	Qwen3.8 27B Alibaba	73±1	api $1.60/M +local	~18GB	2026-08
64	Laguna S 2.1 (118B-A8B) Poolside	73±4	local	~71GB	2026-07
65	Grok 4.20 Multi-Agent (0309) SpaceXAI (xAI)	73±2	api $1.56/M		2026-03
66	Seed2.0 Lite ByteDance Seed	73±4	comparison only
	more from Poolside
149	Laguna M.1 Poolside	67±2	api price unknown +local	~135GB	2026-04
183	Laguna XS 2.1 (33B-A3B) Poolside	66±2	local	~21GB	2026-06
202	Laguna XS.2 Poolside	65±2	local	~21GB	2026-04

see all LLMs · find the ones that fit your machine

$ what it's good at

agent tasks███········█ #1 of 4 * · 93% of its score · 4 tests

test	result	vs the others in its run	difficulty	counts	why
SWE-bench Multilingual lab	78.5% ✎	▁··▁▃▂▅▇█▃▃▁ beats 49/60	hard · avg 28%	▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 66%	small run
SWE-bench Pro lab	59.4% ✎	▅▃▃·▃·▅▅▅█▃▃ beats 13/15	hard · avg 30%	▇▇▇▇▇▇▇▇▇▇▇ 26%	blunt · at its level · small run
Terminal-Bench 2.1 lab	70.2% ✎	▁·▁·▁▁▄▃▄█▃▇ beats 34/56	very hard · avg 7.1%	—	not counted yet: no other model in the same run to compare it with
Toolathlon-Verified lab	49.7% ✎	▂·▂▂▅▃···▂██ beats 6/18	hard · avg 24%	—	not counted yet: no other model in the same run to compare it with

coding▁▁▁▃▆██▇▆▄▃▁ #51 of 469 * · 93% of its score · 2 tests

Beats GPT-5.4, DeepSeek V4 Flash Vision (Exp), Gemini 3.1 Pro (Preview) here, though they rank higher overall. Trails Mistral Large 4 (public preview), Qwen3.7 Plus, Claude Sonnet 4.6 here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
SWE-bench Multilingual lab	78.5% ✎	▁··▁▃▂▅▇█▃▃▁ beats 49/60	hard · avg 28%	▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 66%	small run
SWE-bench Pro lab	59.4% ✎	▅▃▃·▃·▅▅▅█▃▃ beats 13/15	hard · avg 30%	▇▇▇▇▇▇▇▇▇▇▇ 26%	blunt · at its level · small run

$ where to run it

llama.cpp · recommended

version

Install llama.cpp · Homebrew and winget installs name the commands llama-cli and llama-server instead.
macOS / Linux
curl -LsSf https://llama.app/install.sh | sh
Windows
irm https://llama.app/install.ps1 | iex
other ways to install ↗
Run Laguna S 2.1 (118B-A8B) · needs ~96 GB · from huggingface.co/poolside (official)
chat in the terminal
llama cli -hf poolside/Laguna-S-2.1-GGUF:Q4_K_M
Or start a local server and open the address it prints (http://localhost:9931)
llama serve -hf poolside/Laguna-S-2.1-GGUF:Q4_K_M

LM Studio

version

Install LM Studio · Open the app once before using the lms command.
download it (macOS / Windows / Linux) ↗
Run Laguna S 2.1 (118B-A8B) · needs ~96 GB · from huggingface.co/poolside (official)
open in LM Studio ↗
Or download it in the terminal
lms get https://huggingface.co/poolside/Laguna-S-2.1-GGUF@Q4_K_M -y
Then run lms ls to see its name, and lms chat followed by that name

Ollama

version

Install Ollama
download it (macOS / Windows) ↗
Linux
curl -fsSL https://ollama.com/install.sh | sh
Run Laguna S 2.1 (118B-A8B) · needs ~96 GB · from ollama.com
chat in the terminal
ollama run laguna-s-2.1:q4_K_M

MLX (Mac)

version

Install MLX (Mac) · Apple silicon Macs only, text only.
with uv
uv tool install mlx-lm
or
pip install mlx-lm
Run Laguna S 2.1 (118B-A8B) · needs ~66.1 GB, Mac only · from huggingface.co/TensorFold
chat in the terminal
mlx_lm.chat --model TensorFold/Laguna-S-2.1-MLX-4bit
Or start an OpenAI-compatible server at http://localhost:8080/v1
mlx_lm.server --model TensorFold/Laguna-S-2.1-MLX-4bit

Other ways to run it

Run it as published · needs ~235.13 GB, NVIDIA only · from huggingface.co/poolside (official)
vllm serve --model poolside/Laguna-S-2.1 --tensor-parallel-size 4 --tool-call-parser poolside_v1 --reasoning-parser poolside_v1 --enable-auto-tool-choice --served-model-name laguna --default-chat-template-kwargs '{"enable_thinking": true}'

Official weights and documentation: huggingface.co/poolside ↗

$ how the score was worked out · Poolside vendor-reported benc… 93%

Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works

source	results	share of the score	licence
Poolside vendor-reported benchmark scores for its own Laguna models (Hugging Face model cards) · lab-reported	2	▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 93%	Factual vendor-reported score (not licensed content)
prior · pull toward the middle		▇▇ 7%
weighted average		= 73.4

not counted (2)

Poolside vendor-reported benchmark scores for its own Laguna models (Hugging Face model cards)	Terminal-Bench 2.1 [correct]	70.2%	fewer than three models have this test yet, or this source ran it for this model alone
Poolside vendor-reported benchmark scores for its own Laguna models (Hugging Face model cards)	Toolathlon-Verified [correct]	49.7%	fewer than three models have this test yet, or this source ran it for this model alone

$ improve this page

Update model information Add a place to run it Add a test result Propose a benchmark Correct a benchmark score Edit data on GitHub

Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.

data updated 2026-10-11 02:20 UTC · sources and licences · json