Piscina XS.2 needs about 12.9 GB of memory at IQ2_M (a 11.0 GB file) with an 8K context window, so it fits on a 16 GB GPU. This repository has 5 quants, from 7.6 GB (IQ1_M) to 35.6 GB (Q8_0).
Piscina XS.2 – a 33B Mixture‑of‑Experts model quantized for local use
Piscina XS.2 is a 33B total‑parameter Mixture‑of‑Experts model with 3B activated parameters per token. It supports reasoning and tool use and is designed for agentic coding and long‑horizon tasks on a local machine.
Good at
- Reasoning with tool calls
- Long‑horizon agentic coding
- Sliding Window Attention for efficient inference
Limits
- No vision support
- Requires 8k context for memory estimates
- Apache‑2.0 license only for text‑to‑text
Best for
- Local coding assistants
- Long‑form text generation
- Agentic task planning
Who it’s for. Developers and researchers who need a compact, local LLM capable of reasoning and tool integration. The model runs on consumer GPUs and laptops with 8–36 GB of RAM.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| IQ1_M | 7.6 GB | 9.0 GB | Too big | Get file → | |
| — | 8.8 GB | 10.3 GB | Too big | Get file → | |
| IQ2_M | 11.0 GB | 12.9 GB | Too big | Get file → | |
| — | 11.9 GB | 13.9 GB | Too big | Get file → | |
| Q8_0 | 35.6 GB | 40.4 GB | Too big | Get file → |
For 8 GB VRAM, use IQ1_M (file size 7.63 GB, estimated 9 GB memory at 8k context). For 12 GB VRAM, IQ2_M (11.04 GB, 12.9 GB memory). For larger GPUs, Q8_0 (35.6 GB, 40.4 GB memory). Estimates are approximate.
What GPU do you need for Piscina XS.2?
These figures add the file size, an fp16 KV cache at an 8K context window and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.
- 4 GB VRAM: no quant fits at an 8K context.
- 6 GB VRAM: no quant fits at an 8K context.
- 8 GB VRAM: no quant fits at an 8K context.
- 12 GB VRAM: the best quant that fits is the main file (8.8 GB file, about 10.3 GB in use).
- 16 GB VRAM: the best quant that fits is the main file (11.9 GB file, about 13.9 GB in use).
- 24 GB VRAM: the best quant that fits is the main file (11.9 GB file, about 13.9 GB in use).
- 32 GB VRAM: the best quant that fits is the main file (11.9 GB file, about 13.9 GB in use).
- 48 GB VRAM: the best quant that fits is Q8_0 (35.6 GB file, about 40.4 GB in use).
Which Piscina XS.2 quant should you choose?
IQ2_M (11.0 GB) is the middle option in this repository. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.
If memory is tight, the smallest file is IQ1_M at 7.6 GB, with the quality trade-offs that come with fewer bits. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about Piscina XS.2
How much VRAM does Piscina XS.2 need?
About 12.9 GB at IQ2_M with an 8K context window. That is the 11.0 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.
Can I run Piscina XS.2 on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 9.0 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Piscina XS.2 GGUF?
It depends on the quant: IQ1_M is 7.6 GB, file is 8.8 GB, IQ2_M is 11.0 GB, file is 11.9 GB, Q8_0 is 35.6 GB.
What context length does Piscina XS.2 support?
The GGUF header lists a trained context length of 262,144 tokens (256K).
Where can I download Piscina XS.2 GGUF?
The files are on Hugging Face in the poolside-laguna-hackathon/Piscina-XS.2-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does Piscina XS.2 use?
Hugging Face lists the license as apache-2.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.
Can I run Piscina XS.2 on a GPU with 8 GB VRAM?
IQ1_M requires about 9 GB of memory at 8k context, so 8 GB VRAM may be insufficient for stable inference.
Is Piscina XS.2 compatible with macOS?
Yes, it can run on a Mac with 36 GB of RAM, as the 33B model fits within that memory budget.
Does Piscina XS.2 support vision or multimodal inputs?
No, it is a text‑to‑text model and does not support vision modalities.