KA

KAT-Dev

KatGGUF by MungertUpdated 2025-10-05
Download on Hugging Face
Parameters32.8B
Context128K
Layers64
Architectureqwen3
Smallest quant0.0 GB
Quick answer

KAT-Dev needs about 22.8 GB of memory at Q4_K_M (a 20.2 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 26 quants, from 0.0 GB to 47.0 GB (F16). Longer context needs more: at 32K tokens the KV cache alone adds about 8.6 GB.

KAT-Dev 32B: a Qwen3-based model for software engineering tasks

KAT-Dev is a 32.8B‑parameter Qwen3 model optimized for software engineering. It was trained with mid‑training, supervised fine‑tuning, reinforcement fine‑tuning, and large‑scale agentic reinforcement learning. The model is available in multiple GGUF quantizations.

Good at

  • Software engineering code generation
  • Instruction following
  • Multi‑turn interaction

Limits

  • No vision support
  • Large VRAM requirement for higher precision quants
  • License is other

Best for

  • Code completion in IDEs
  • Automated code review tools
  • SWE‑bench style evaluations

Who it’s for. Developers and researchers needing a large language model focused on software engineering. The model is open‑source and can be fine‑tuned for specialized coding tasks.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
—0.0 GB 2.7 GB
Fits Get file →
IQ1_S8.9 GB 11.6 GB
Too big Get file →
IQ1_M9.9 GB 12.5 GB
Too big Get file →
IQ2_XXS10.2 GB 12.8 GB
Too big Get file →
IQ2_XS11.1 GB 13.7 GB
Too big Get file →
IQ2_S11.4 GB 14.1 GB
Too big Get file →
IQ2_M12.0 GB 14.7 GB
Too big Get file →
Q2_K_S12.2 GB 14.8 GB
Too big Get file →
Q2_K_M12.4 GB 15.0 GB
Too big Get file →
IQ3_XXS13.8 GB 16.4 GB
Too big Get file →
IQ3_XS14.2 GB 16.8 GB
Too big Get file →
IQ3_M15.7 GB 18.3 GB
Too big Get file →
Q3_K_S15.9 GB 18.6 GB
Too big Get file →
Q3_K_M16.1 GB 18.8 GB
Too big Get file →
IQ4_XS17.9 GB 20.6 GB
Too big Get file →
IQ4_NL18.5 GB 21.1 GB
Too big Get file →
Q4_K_S19.2 GB 21.8 GB
Too big Get file →
Q4_019.2 GB 21.9 GB
Too big Get file →
Q4_K_M20.2 GB 22.8 GB
Too big Get file →
Q4_120.6 GB 23.2 GB
Too big Get file →
Q5_023.1 GB 25.8 GB
Too big Get file →
Q5_K_M23.8 GB 26.4 GB
Too big Get file →
Q5_125.1 GB 27.7 GB
Too big Get file →
Q6_K_M27.3 GB 29.9 GB
Too big Get file →
Q8_034.8 GB 37.5 GB
Too big Get file →
F1647.0 GB 49.6 GB
Too big Get file →

For 8 GB VRAM, use IQ1_S (8.93 GB file, 11.6 GB memory). For 12 GB VRAM, use Q2_K_M (12.38 GB file, 15 GB memory). For 16 GB VRAM, use Q3_K_M (16.14 GB file, 18.8 GB memory). Estimates are approximate.

What GPU do you need for KAT-Dev?

These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) 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 the main file (0.0 GB file, about 2.7 GB in use).
  • 6 GB VRAM: the best quant that fits is the main file (0.0 GB file, about 2.7 GB in use).
  • 8 GB VRAM: the best quant that fits is the main file (0.0 GB file, about 2.7 GB in use).
  • 12 GB VRAM: the best quant that fits is IQ1_S (8.9 GB file, about 11.6 GB in use).
  • 16 GB VRAM: the best quant that fits is Q2_K_M (12.4 GB file, about 15.0 GB in use).
  • 24 GB VRAM: the best quant that fits is Q4_1 (20.6 GB file, about 23.2 GB in use).
  • 32 GB VRAM: the best quant that fits is Q6_K_M (27.3 GB file, about 29.9 GB in use).
  • 48 GB VRAM: the best quant that fits is Q8_0 (34.8 GB file, about 37.5 GB in use).

Which KAT-Dev quant should you choose?

Q4_K_M is the usual starting point for KAT-Dev: it balances file size and quality well. Here it is a 20.2 GB file.

The Q8_0 file is 34.8 GB, about 1.7 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.

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

Architectureqwen3
Parameters32.8B
Layers64
Trained context131,072 tokens
KV cache (fp16)256 KB per token

Source

Quantized byMungert
FamilyKat
Licenseother
Downloads (HF)765
Last updated2025-10-05

Frequently asked questions about KAT-Dev

How much VRAM does KAT-Dev need?

About 22.8 GB at Q4_K_M with an 8K context window. That is the 20.2 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 KAT-Dev on an 8 GB GPU?

Yes, with the smaller files: the 0.0 GB file needs about 2.7 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of KAT-Dev GGUF?

It depends on the quant: file is 0.0 GB, IQ1_S is 8.9 GB, IQ1_M is 9.9 GB, IQ2_XXS is 10.2 GB, IQ2_XS is 11.1 GB, IQ2_S is 11.4 GB and 20 more.

What context length does KAT-Dev support?

The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 2.1 GB at 8K and 8.6 GB at 32K.

Where can I download KAT-Dev GGUF?

The files are on Hugging Face in the Mungert/KAT-Dev-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 KAT-Dev 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 KAT-Dev on 8 GB VRAM?

Yes, using the IQ1_S quantization which requires about 11.6 GB memory at 8 k context.

Does KAT-Dev support vision tasks?

No, the model has no vision capability.

What is the license for KAT-Dev?

The model is released under an other license; check the repository for details.

Similar models

Browse all →

More from Mungert

All by Mungert →

More Kat models

All Kat →

Other medium models

All medium models →