Laguna XS 2.1
Summary
Laguna XS 2.1 is an open-weight language model from Poolside, released on 2 Jul 2026. It accepts text and generates text, with a 262K-token context window and up to 16K 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 19 Nov 2026. Every figure on this page links to its source.
Specifications
| Developer | Poolside |
|---|---|
| Release date | 2 Jul 2026 [models.dev] |
| Status | Generally available [models.dev] |
| Weights | Open weights [Hugging Face] |
| License | openmdw-1.1 [Hugging Face] |
| Parameters | 33.4B [Hugging Face] |
| Context window | 262K (262,144 tokens) [models.dev] |
| Max output | 16K tokens [models.dev] |
| Input | Text [models.dev] |
| Output | Text [models.dev] |
| Reasoning mode | Yes [models.dev] |
| Tool calling | Yes [models.dev] |
| Structured output | No [models.dev] |
| Retirement date | 31 Oct 2026 [OpenRouter] |
| Hugging Face | poolside/Laguna-XS-2.1 [Hugging Face] |
Laguna XS 2.1 API pricing by provider
USD per million tokens. Sorted by input price.
| Provider | Input | Output | Cached input | Context | Source |
|---|---|---|---|---|---|
| NVIDIA NIM poolside/laguna-xs-2.1 | Free | Free | – | 262K | models.dev, 19 Nov 2026 |
Running Laguna XS 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.
| Quantization | Weights | Total, 8K context | Total, 32K context |
|---|---|---|---|
| Q4_K_M (4-bit) | 20 GB | 21 GB | 22 GB |
| Q8_0 (8-bit) | 36 GB | 37 GB | 38 GB |
| FP16 / BF16 (16-bit) | 67 GB | 68 GB | 69 GB |
At 4-bit with an 8K conversation, Laguna XS 2.1 needs about 21 GB, so it fits on a graphics card with 24 GB or more, such as the NVIDIA GeForce RTX 3090 24 GB. Check your own hardware, or see Best GPU for local LLMs on our sister site AIPCs.
Frequently asked questions
What is the context window of Laguna XS 2.1?
Laguna XS 2.1 has a context window of 262,144 tokens (262K) and can generate up to 16,384 tokens in a single response, according to models.dev.
Is Laguna XS 2.1 open source?
Laguna XS 2.1 is an open-weight model: its weights are published on Hugging Face as poolside/Laguna-XS-2.1 under the openmdw-1.1 license. Check the license terms before commercial use.
When was Laguna XS 2.1 released?
Poolside released Laguna XS 2.1 on 2 Jul 2026, according to models.dev.
How much VRAM does Laguna XS 2.1 need?
About 21 GB at 4-bit (Q4_K_M) with an 8K-token conversation, and about 37 GB at 8-bit. The 4-bit model fits on a graphics card with 24 GB or more, such as the NVIDIA GeForce RTX 3090 24 GB. 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.
| Model | Lab | Released | Context | Input $/1M | Output $/1M | Weights |
|---|---|---|---|---|---|---|
| Hy3 | Tencent (Hunyuan) | 6 Jul 2026 | 262K | $0.13 | $0.53 | Open |
| Ornith-1.0-35B-FP8 | Deepreinforce | 25 Jun 2026 | 262K | – | – | Open |
| KAT-Coder-Pro V2.5 | Kwaipilot | 10 Jul 2026 | 262K | – | – | – |
| Macaron V1 Tall | Mindlab | 21 Jul 2026 | 262K | $0.25 | $1.43 | Open |
| Kimi K2.7 Code | Moonshot AI | 12 Jun 2026 | 262K | $0.60 | $3.50 | Open |
| Kimi K2.7 Code High Speed | Moonshot AI | 12 Jun 2026 | 262K | $1.90 | $8 | Open |
Change history
- No longer listed on openrouter
- Max output changed from 33K to 16K
- Name changed from Laguna XS 2.1 (free) to Laguna XS 2.1