Poolside

Laguna S 2.1

Generally availableOpen weights1M contextReasoning

Summary

Laguna S 2.1 is an open-weight language model from Poolside, released on 21 Jul 2026. It accepts text and generates text, with a 1M-token context window and up to 33K output tokens per response. It is scheduled for retirement on 31 Oct 2026.

Agentic coding model from Poolside in the XS size class for local deployment

Last verified 18 Nov 2026. Every figure on this page links to its source.

Specifications

DeveloperPoolside
Release date21 Jul 2026 [models.dev]
StatusGenerally available [models.dev]
WeightsOpen weights [Hugging Face]
Licenseopenmdw-1.1 [Hugging Face]
Parameters118B [Hugging Face]
Context window1M (1,048,576 tokens) [models.dev]
Max output33K tokens [models.dev]
InputText [models.dev]
OutputText [models.dev]
Reasoning modeYes [models.dev]
Tool callingYes [models.dev]
Structured outputNo [models.dev]
Retirement date31 Oct 2026 [OpenRouter]
Hugging Facepoolside/Laguna-S-2.1 [Hugging Face]

Laguna S 2.1 API pricing by provider

USD per million tokens. Sorted by input price.

We do not currently track a paid API listing for this model.

Running Laguna S 2.1 locally

Estimated memory, from the parameter count and the architecture in config.json on Hugging Face. Includes about 0.8 GB of runtime overhead.

QuantizationWeightsTotal, 8K contextTotal, 32K context
Q4_K_M (4-bit)71 GB73 GB74 GB
Q8_0 (8-bit)125 GB126 GB127 GB
FP16 / BF16 (16-bit)235 GB236 GB238 GB

At 4-bit with an 8K conversation, Laguna S 2.1 needs about 73 GB, more than any single consumer graphics card; it needs a unified-memory machine, a multi-GPU setup or a split between graphics card and system RAM. Check your own hardware, or see Best computer for local AI on our sister site AIPCs.

Frequently asked questions

What is the context window of Laguna S 2.1?

Laguna S 2.1 has a context window of 1,048,576 tokens (1M) and can generate up to 32,768 tokens in a single response, according to models.dev.

Is Laguna S 2.1 open source?

Laguna S 2.1 is an open-weight model: its weights are published on Hugging Face as poolside/Laguna-S-2.1 under the openmdw-1.1 license. Check the license terms before commercial use.

When was Laguna S 2.1 released?

Poolside released Laguna S 2.1 on 21 Jul 2026, according to models.dev.

How much VRAM does Laguna S 2.1 need?

About 73 GB at 4-bit (Q4_K_M) with an 8K-token conversation, and about 126 GB at 8-bit. That is more than any single consumer graphics card holds. These are estimates from the model's published parameter count and architecture.

Similar models from other labs

Comparable on output type, weights, release date, price and context window. Prices are the lowest listed, USD per million tokens.

ModelLabReleasedContextInput $/1MOutput $/1MWeights
Macaron V1 VentiMindlab21 Jul 20261M$0.82$2.48Open
Kimi K3Moonshot AI16 Jul 20261M$0.19$10Open
Inkling SmallThinkingmachines30 Jul 20261M$0.45$1.20Open
DeepSeek V4 Flash 0731DeepSeek31 Jul 20261M$0.06$0.18Open
LongCat-2.0Meituan30 Jun 20261M$0.30$1.20Open
DeepSeek V4 ProDeepSeek13 Aug 20261M$0.11$0.87Open

Change history

  • No longer listed on vercel
  • Max output changed from 131K to 33K
  • Context window changed from 1M to 1M
  • No longer listed on openrouter
  • No longer listed on vercel
  • Context window changed from 1M to 1M
  • Max output changed from 33K to 131K
  • Context window changed from 262K to 1M
  • Name changed from Laguna S 2.1 (free) to Laguna S 2.1
  • vercel price changed: input $0.10 → $0.09, output $0.20 → $0.18 per 1M tokens

Sources