Can I run Laguna XS 2.1 locally?

Yes, the weights are open under the OpenMDW-1.1 licence and the model is available on Ollama and llama.cpp. Poolside says a Mac with 36 GB of RAM is sufficient.

What does Laguna XS 2.1 cost, and how long is the context?

Via API it is $0.06 per million input tokens and $0.12 per million output tokens, with a 262,144-token context window and up to 32K output tokens per request.

How does Laguna XS 2.1 differ from Laguna XS.2?

It improves on Laguna XS.2 with a +5.4% gain on SWE-bench Multilingual and stronger scores on terminal-style tasks, along with FP8 KV cache quantization for lower memory use.

Laguna XS 2.1

2460°

#llm #poolside #reasoning #coding #agents #long context

Run this model

Offers / Ads

OpenRouter Terminal / API

Pay per token

Chat with this model in the browser or call it from your code with one API key.

Poolside | 262,144 token context | $0.06/M input, $0.12/M output. Laguna XS 2.1 is the latest coding agent model in the 33B-A3B category from PoolsideOpens in new tab and a step forward from their Laguna XS.2 model (released in April 2026). It combines tool calling and reasoning capabilities with a compact footprint, offering a 256K context window and up to 32K output tokens. Quantized to FP8 for fast, cost-efficient agentic coding workflows. Laguna XS 2.1 is designed for software engineering and agentic coding use cases, and you are responsible for confirming that it is appropriate for your…Poolside | 262,144 token context | $0.06/M input, $0.12/M output. Laguna XS 2.1 is the latest coding agent model in the 33B-A3B category from PoolsideOpens in new tab and a step forward from their Laguna XS.2 model (released in April 2026). It combines tool calling and reasoning capabilities with a compact footprint, offering a 256K context window and up to 32K output tokens. Quantized to FP8 for fast, cost-efficient agentic coding workflows. Laguna XS 2.1 is designed for software engineering and agentic coding use cases, and you are responsible for confirming that it is appropriate for your intended application. Laguna XS 2.1 is subject to the OpenMDW-1.1 LicenseOpens in new tab, and should be used consistently with Poolside's Acceptable Use PolicyOpens in new tab. We advise against circumventing Laguna XS 2.1 safety guardrails without implementing substantially equivalent mitigations appropriate for your use case. Please report security vulnerabilities or safety concerns to [email protected]. If you are using Laguna XS 2.1 for free, we may use your inputs and outputs to train and improve our models.

Guides for this model

Ollama vs LM Studio (2026): which local AI app to use, and which top models each one runs

Best local LLM by use case (October 2026): the right open model for every job

How much VRAM do I actually need for local AI?

What is a GGUF file? Quantized models explained (Q4, Q8, fp8, NF4)

All guides

Captain's notes

Written by Captain, the house editor

Notes from 2026-10-13

Laguna XS 2.1 is a 33B Mixture-of-Experts coding agent from Poolside, with only 3B parameters activated per token, making it genuinely runnable on a 36 GB Mac. It pairs native reasoning with tool calling across a 256K context window, and its open weights ship under a fully permissive commercial licence. If your work is agentic coding or long-horizon software engineering tasks, this is built for exactly that.

Known for

Agentic coding with interleaved reasoning between tool calls
Runs locally on a 36 GB Mac via Ollama or llama.cpp
256K context window with up to 32K output tokens
Fully permissive OpenMDW-1.1 licence, commercial use included
FP8-quantized KV cache keeps memory and inference costs low

Watch out for

Free-tier usage allows Poolside to train on your inputs and outputs.
Benchmark scores trail Qwen3.6-35B-A3B on every published coding metric despite similar size.
Designed for software engineering; suitability for other use cases is your call to confirm.

Best for

Prompts to try

Debug a failing test suite

I have a Python Flask API. The test suite passes locally but fails in CI with this error: AssertionError: 401 != 200 on the /api/orders endpoint. The endpoint requires a Bearer token. Here is the test file and the route handler. Identify the root cause and rewrite the test so it passes correctly in both environments.

Refactor for async agents

Refactor this synchronous Python script that calls three external APIs in sequence into an async version using asyncio and aiohttp. The three calls are independent of each other. Preserve all error handling, add a timeout of 10 seconds per request, and explain each change you make.

Plan a multi-step code migration

I need to migrate a Node.js Express app from CommonJS require() to ES module import/export syntax. The project has 40 files, uses Jest for tests, and has a few dynamic require() calls. Give me a step-by-step migration plan, flag the dynamic require cases separately, and write the updated package.json and Jest config.

Compare with

Laguna S 2.1
Open-weights model
1,048,576 token context · $0.09/M input, $0.18/M output
GPT-6 Luna
API-only model
1,050,000 token context · $0.10/M input, $0.50/M output
Claude Opus 5.5
API-only model
1,000,000 token context · $4.00/M input, $20.00/M output
Muse Spark 1.3 Contributor
API-only model
1,048,576 token context · $0.10/M input, $0.20/M output

More text models from poolside

Questions people ask

Can I run Laguna XS 2.1 locally?: Yes, the weights are open under the OpenMDW-1.1 licence and the model is available on Ollama and llama.cpp. Poolside says a Mac with 36 GB of RAM is sufficient.
What does Laguna XS 2.1 cost, and how long is the context?: Via API it is $0.06 per million input tokens and $0.12 per million output tokens, with a 262,144-token context window and up to 32K output tokens per request.
How does Laguna XS 2.1 differ from Laguna XS.2?: It improves on Laguna XS.2 with a +5.4% gain on SWE-bench Multilingual and stronger scores on terminal-style tasks, along with FP8 KV cache quantization for lower memory use.

Captain's verdict
A lean, locally-runnable coding agent with a permissive licence and real benchmark gains, though Qwen3.6-35B-A3B still leads it on every published score.

Model card on Hugging Face

Captain writes these notes from the model's own description and model card. Prices and context sizes are the provider's at the time of writing and can change.

Ranking badge

Updated Oct 22, 2026

Made this model? Show this badge on your page.

Paste it into your Hugging Face card or Civitai description.

Change colors / personalize / get code

Not this creator?

Get your own badge

How this score adds up

This model has no popularity data from any source yet, so it carries a stand-in number below 7,000° that keeps the lists complete. It never ranks above a model with data. The breakdown appears here as soon as a source reports it.

Glossary: the terms used here

Heat score: One number for how much a model is used and wanted right now, from 0° to 9,999°. Every list on this site that says hottest is sorted by it and nothing else.
Floor: The 7,000° every model with real data starts from. The scale runs from 0° to 9,999°; the part below the floor is reserved for stand-ins (models without data), so that a model with data always ranks above one without. The floor is the same for everyone and decides nothing; the degrees above it are the measurement.
Download-equivalents: Everything the score treats as a download, added up: downloads themselves, each torrent seeder as 1,000, each AI Horde image and each run on a generation site as one.
Log scale: Counting by tens, not by ones: going from 1,000 to 10,000 downloads adds the same degrees as going from 10,000 to 100,000. It keeps giants from crowding out everything else.
Likes: Thumbs up and likes from every source, worth half a download each on the log scale.
Adoption rate: How many downloads a day a model gets right now, from the difference between two readings of a source. A quarter of the weight of the download term.
Measured and estimated: Measured means two readings of a source exist and the difference is real. Estimated means only today's total is known and the past is spread over the model's life; the site says which one it is showing.
Seeder: Someone who keeps a copy of the weights online for others to download over a torrent, or a hosted mirror doing the same. Each one counts as 1,000 downloads.
AI Horde: A volunteer network that generates images for free on donated computers. The number of images it has made with a model is public, and each counts as one download.
Generation sites: Sites that host their own copies of a model and let people generate with it, such as PixAI, SeaArt and Yodayo. Their downloads count like Civitai's and each generation run as one download.
OpenRouter tokens: A text model served through an API, such as Claude or GPT, has no weights to download, so its usage is measured in tokens: the pieces of text it read and wrote through OpenRouter, a gateway many apps use, over the last 30 days. Every 10,000,000 tokens count as one download. Source: OpenRouter (openrouter.ai/rankings), CC BY 4.0.
Full model (checkpoint): The weights themselves, the file that generates on its own. It gets a bonus because add-ons depend on it.
LoRA and embedding: Small add-on files that steer someone else's weights toward a style, a character or a concept. They cannot run alone.
Family and weights: The base a model is built on, such as SD 1.5, SDXL, Flux or Wan. Every model inherits the release date of its family's weights.
Age: Degrees taken off for how long ago the family's weights came out: 150° a year, up to 750°. It keeps old downloads that had years to pile up from freezing the top of the list.
Family boost: A small, bounded nudge for LoRAs of a family this site is pushing forward right now, always listed in the breakdown.
Stand-in score: A stable number below the floor given to a model no source reports on yet, so lists stay complete. It never ranks above a model with data.
New! (grace period): A model less than 7 days old shows New! instead of a score and holds no place in the rankings. Its score is computed from day one, but a model one day old grows faster than anything, so without the wait the week's #1 would always be whatever came out yesterday rather than what held up for a week.
Trend: How a model's growth over a period compares with the typical model's growth over the same period. Faster than typical reads as a rise, slower as a fall.
Typical model: The middle of the catalog: half the models with a figure grew faster over the period, half slower.

Similar Models

AI Models for Image, Video & Text Generation

Laguna XS 2.1 is a text model for the Text / LLM family listed on ModelTrends.ai, a read-only catalog of open source AI image, video and text models. Compare it with other Text / LLM models and check its heat score to see how it is trending.

More Text / LLM models llm poolside reasoning coding agents long context local

Browse by Model Family