GE

Gemma 3 27b It PaperWitch Heresy

GemmaGGUF by mradermacherUpdated 2026-07-25
Download on Hugging Face
Parameters27B
Context128K
Layers62
Architecturegemma3
Smallest quant0.0 GB
Quick answer

Gemma 3 27b It PaperWitch Heresy needs about 21.2 GB of memory at Q4_K_M (a 16.5 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 24 quants, from 0.0 GB to 22.2 GB (Q6_K). Longer context needs more: at 32K tokens the KV cache alone adds about 16.6 GB.

Gemma 3 27b It PaperWitch Heresy – a 27 billion‑parameter instruction‑tuned vision‑capable model

Gemma 3 27b It PaperWitch Heresy is a 27 billion‑parameter model fine‑tuned by P‑E‑W. It supports text and image input with a 128K context window. The model is available in a range of GGUF quantizations.

Good at

  • Multimodal text and image tasks
  • Large context handling
  • Instruction‑driven generation

Limits

  • No vision support in the provided GGUF files
  • License is gemma – check usage terms
  • Requires large VRAM for higher‑quality quants

Best for

  • Chatbots needing long context
  • Document summarization with images
  • Multilingual Q&A

Who it’s for. Developers and researchers who need a large, instruction‑tuned model with multimodal capabilities. Suitable for deployment on systems with sufficient GPU memory.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
—0.0 GB 4.7 GB
Fits Get file →
IQ1_S6.3 GB 10.9 GB
Too big Get file →
IQ1_M6.8 GB 11.5 GB
Too big Get file →
IQ2_XXS7.7 GB 12.3 GB
Too big Get file →
IQ2_XS8.4 GB 13.1 GB
Too big Get file →
IQ2_S8.8 GB 13.4 GB
Too big Get file →
IQ2_M9.5 GB 14.2 GB
Too big Get file →
Q2_K_S9.8 GB 14.4 GB
Too big Get file →
Q2_K10.5 GB 15.2 GB
Too big Get file →
IQ3_XXS10.7 GB 15.4 GB
Too big Get file →
IQ3_XS11.6 GB 16.2 GB
Too big Get file →
IQ3_S12.2 GB 16.8 GB
Too big Get file →
Q3_K_S12.2 GB 16.8 GB
Too big Get file →
IQ3_M12.5 GB 17.2 GB
Too big Get file →
Q3_K_M13.4 GB 18.1 GB
Too big Get file →
Q3_K_L14.5 GB 19.2 GB
Too big Get file →
IQ4_XS14.8 GB 19.4 GB
Too big Get file →
Q4_015.6 GB 20.3 GB
Too big Get file →
Q4_K_S15.7 GB 20.3 GB
Too big Get file →
Q4_K_M16.5 GB 21.2 GB
Too big Get file →
Q4_117.2 GB 21.8 GB
Too big Get file →
Q5_K_S18.8 GB 23.4 GB
Too big Get file →
Q5_K_M19.3 GB 23.9 GB
Too big Get file →
Q6_K22.2 GB 26.8 GB
Too big Get file →

For 8 GB VRAM, use IQ1_S (6.26 GB) or IQ1_M (6.8 GB). For 12 GB VRAM, Q2_XXS (7.69 GB) or Q2_XS (8.44 GB) are good choices. Higher quants like Q4_K_M (16.55 GB) require 20 GB+ VRAM.

What GPU do you need for Gemma 3 27b It PaperWitch Heresy?

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: no quant fits at an 8K context.
  • 6 GB VRAM: the best quant that fits is the main file (0.0 GB file, about 4.7 GB in use).
  • 8 GB VRAM: the best quant that fits is the main file (0.0 GB file, about 4.7 GB in use).
  • 12 GB VRAM: the best quant that fits is IQ1_M (6.8 GB file, about 11.5 GB in use).
  • 16 GB VRAM: the best quant that fits is IQ3_XXS (10.7 GB file, about 15.4 GB in use).
  • 24 GB VRAM: the best quant that fits is Q5_K_M (19.3 GB file, about 23.9 GB in use).
  • 32 GB VRAM: the best quant that fits is Q6_K (22.2 GB file, about 26.8 GB in use).

Which Gemma 3 27b It PaperWitch Heresy quant should you choose?

Q4_K_M is the usual starting point for Gemma 3 27b It PaperWitch Heresy: it balances file size and quality well. Here it is a 16.5 GB file.

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

Architecturegemma3
Parameters27B
Layers62
Trained context131,072 tokens
KV cache (fp16)496 KB per token

Source

Quantized bymradermacher
FamilyGemma
Licensegemma
Downloads (HF)733
Last updated2026-07-25

Frequently asked questions about Gemma 3 27b It PaperWitch Heresy

How much VRAM does Gemma 3 27b It PaperWitch Heresy need?

About 21.2 GB at Q4_K_M with an 8K context window. That is the 16.5 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 Gemma 3 27b It PaperWitch Heresy on an 8 GB GPU?

Yes, with the smaller files: the 0.0 GB file needs about 4.7 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of Gemma 3 27b It PaperWitch Heresy GGUF?

It depends on the quant: file is 0.0 GB, IQ1_S is 6.3 GB, IQ1_M is 6.8 GB, IQ2_XXS is 7.7 GB, IQ2_XS is 8.4 GB, IQ2_S is 8.8 GB and 18 more.

What context length does Gemma 3 27b It PaperWitch Heresy support?

The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 4.2 GB at 8K and 16.6 GB at 32K.

Where can I download Gemma 3 27b It PaperWitch Heresy GGUF?

The files are on Hugging Face in the mradermacher/gemma-3-27b-it-PaperWitch-heresy-i1-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 Gemma 3 27b It PaperWitch Heresy use?

Hugging Face lists the license as gemma. 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 this model on 8 GB VRAM?

Use the IQ1_S or IQ1_M quant, each under 8 GB file size, but memory usage will be higher than the file size.

Does this model support image input?

Yes, it is a vision model; image input is encoded to 256 tokens at 896×896 resolution.

What license applies to this model?

The model is licensed under the gemma license; review the terms for usage restrictions.

Other GGUFs of gemma-3-27b-it-PaperWitch-heresy

Similar models

Browse all →

More from mradermacher

All by mradermacher →

More Gemma models

All Gemma →

Other medium models

All medium models →