QW

Katy 27B

QwenGGUF by sirfyynUpdated 2026-10-22
Download on Hugging Face
Parameters1.92B
Context256K
Layers—
Architecturedflash
Smallest quant0.0 GB
Quick answer

Katy 27B needs about 11.5 GB of memory at IQ2_S (a 9.8 GB file) with an 8K context window, so it fits on a 12 GB GPU. This repository has 5 quants, from 0.0 GB to 15.3 GB (IQ4_XS).

Katy 27B – a Qwen3.8‑27B derivative in GGUF format

Katy 27B is a quantized version of Swift‑1.5‑Qwen3.8‑27B. It uses the dflash architecture and includes an importance matrix for drafting. The model is available in several quantizations for llama.cpp.

Good at

  • Reasoning tasks
  • Agentic workflows
  • Coding prompts

Limits

  • No vision support
  • License is other
  • Context limited to 262144 tokens

Best for

  • Local inference on GPUs
  • Fine‑tuning with llama.cpp
  • Low‑latency chat

Who it’s for. Developers and researchers who need a compact, efficient Qwen derivative for local deployment. The model is suitable for those who can run llama.cpp and want to experiment with drafting and importance matrices.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
—0.0 GB 0.5 GB
Fits Get file →
Q8_02.1 GB 2.8 GB
Fits Get file →
IQ2_S9.8 GB 11.5 GB
Too big Get file →
IQ3_S12.6 GB 14.6 GB
Too big Get file →
IQ4_XS15.3 GB 17.6 GB
Too big Get file →

For 8 GB VRAM, use Q8_0 (2.06 GB file, 2.8 GB memory). For 4 GB VRAM, use IQ2_S (9.81 GB file, 11.5 GB memory) or IQ3_S (12.6 GB file, 14.6 GB memory). For 2 GB VRAM, use IQ4_XS (15.31 GB file, 17.6 GB memory). Estimates are approximate.

What GPU do you need for Katy 27B?

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: the best quant that fits is Q8_0 (2.1 GB file, about 2.8 GB in use).
  • 6 GB VRAM: the best quant that fits is Q8_0 (2.1 GB file, about 2.8 GB in use).
  • 8 GB VRAM: the best quant that fits is Q8_0 (2.1 GB file, about 2.8 GB in use).
  • 12 GB VRAM: the best quant that fits is IQ2_S (9.8 GB file, about 11.5 GB in use).
  • 16 GB VRAM: the best quant that fits is IQ3_S (12.6 GB file, about 14.6 GB in use).
  • 24 GB VRAM: the best quant that fits is IQ4_XS (15.3 GB file, about 17.6 GB in use).

Which Katy 27B quant should you choose?

IQ2_S (9.8 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 the first one in the table at 0.0 GB, with the quality trade-offs that come with fewer bits.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturedflash
Parameters1.92B
Trained context262,144 tokens

Source

Quantized bysirfyyn
FamilyQwen
Licenseother
Downloads (HF)2,119
Last updated2026-10-22

Frequently asked questions about Katy 27B

How much VRAM does Katy 27B need?

About 11.5 GB at IQ2_S with an 8K context window. That is the 9.8 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 Katy 27B on an 8 GB GPU?

Yes, with Q8_0: the 2.1 GB file needs about 2.8 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of Katy 27B GGUF?

It depends on the quant: file is 0.0 GB, Q8_0 is 2.1 GB, IQ2_S is 9.8 GB, IQ3_S is 12.6 GB, IQ4_XS is 15.3 GB.

What context length does Katy 27B support?

The GGUF header lists a trained context length of 262,144 tokens (256K).

Where can I download Katy 27B GGUF?

The files are on Hugging Face in the sirfyyn/Katy-27B-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 Katy 27B use?

Hugging Face lists the license as other. 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 Katy 27B on 8 GB VRAM?

Yes, the Q8_0 quantization requires about 2.8 GB of VRAM at 8k context.

Does Katy 27B support vision?

No, the model has no vision capabilities.

What is the maximum context length?

The model supports up to 262144 tokens.

Other GGUFs of Swift-1.5-Qwen3.8-27b

Similar models

Browse all →

More from sirfyyn

All by sirfyyn →

More Qwen models

All Qwen →

Other tiny models

All tiny models →