GE

gemma-2-27b-it

GemmaGGUF by bartowskiUpdated 2024-08-03
Download on Hugging Face
Parameters27.2B
Context8K
Layers46
Architecturegemma2
Smallest quant8.4 GB
Quick answer

gemma-2-27b-it needs about 20.2 GB of memory at Q4_K_M (a 16.6 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 25 quants, from 8.4 GB (IQ2_XS) to 108.9 GB (F32).

Gemma‑2‑27B‑IT quantized for llama.cpp

Gemma‑2‑27B‑IT is a 27‑billion‑parameter language model from the Gemma family. It supports 8192‑token context and is available in multiple GGUF quantizations. The model is released under the gemma license and has no system prompt support.

Good at

  • Large‑context text generation
  • Multilingual language tasks
  • Fine‑tuned instruction following

Limits

  • No vision support
  • Requires 8‑k context for estimates
  • License restricts commercial use

Best for

  • Chatbot back‑ends
  • Research prototyping
  • Educational demos

Who it’s for. Developers and researchers who need a large‑parameter model that can run on llama.cpp. The model is suitable for those who can manage the memory requirements and comply with the gemma license.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ2_XS8.4 GB 12.0 GB
Too big Get file →
IQ2_S8.7 GB 12.2 GB
Too big Get file →
IQ2_M9.4 GB 13.0 GB
Too big Get file →
Q2_K10.4 GB 14.0 GB
Too big Get file →
Q2_K_L10.7 GB 14.3 GB
Too big Get file →
IQ3_XXS10.8 GB 14.3 GB
Too big Get file →
IQ3_XS11.6 GB 15.1 GB
Too big Get file →
Q3_K_S12.2 GB 15.8 GB
Too big Get file →
IQ3_M12.5 GB 16.0 GB
Too big Get file →
Q3_K_M13.4 GB 17.0 GB
Too big Get file →
Q3_K_L14.5 GB 18.1 GB
Too big Get file →
Q3_K_XL14.8 GB 18.4 GB
Too big Get file →
IQ4_XS14.8 GB 18.4 GB
Too big Get file →
Q4_K_S15.7 GB 19.3 GB
Too big Get file →
Q4_K_M16.6 GB 20.2 GB
Too big Get file →
Q4_K_L16.9 GB 20.5 GB
Too big Get file →
Q5_K_S18.9 GB 22.5 GB
Too big Get file →
Q5_K_M19.4 GB 23.0 GB
Too big Get file →
Q5_K_L19.7 GB 23.3 GB
Too big Get file →
Q6_K22.3 GB 25.9 GB
Too big Get file →
Q6_K_L22.6 GB 26.2 GB
Too big Get file →
Q8_028.9 GB 32.5 GB
Too big Get file →
Q8_030.0 GB 33.6 GB
Too big Get file →
F32108.9 GB (3 parts) 112.5 GB
Too big Get file →
F32108.9 GB (3 parts) 112.5 GB
Too big Get file →

For 8 GB VRAM, use Q4_K_S (15.74 GB file, ~19.3 GB VRAM). For 12 GB VRAM, Q5_K_S (18.88 GB file, ~22.5 GB VRAM). For 16 GB VRAM, Q5_K_M (19.41 GB file, ~23 GB VRAM). Memory estimates are approximate.

What GPU do you need for gemma-2-27b-it?

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: 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 IQ2_XS (8.4 GB file, about 12.0 GB in use).
  • 16 GB VRAM: the best quant that fits is IQ3_M (12.5 GB file, about 16.0 GB in use).
  • 24 GB VRAM: the best quant that fits is Q5_K_L (19.7 GB file, about 23.3 GB in use).
  • 32 GB VRAM: the best quant that fits is Q6_K_L (22.6 GB file, about 26.2 GB in use).
  • 48 GB VRAM: the best quant that fits is Q8_0 (30.0 GB file, about 33.6 GB in use).

Which gemma-2-27b-it quant should you choose?

Q4_K_M is the usual starting point for gemma-2-27b-it: it balances file size and quality well. Here it is a 16.6 GB file.

The Q8_0 file is 28.9 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 IQ2_XS at 8.4 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

Architecturegemma2
Parameters27.2B
Layers46
Trained context8,192 tokens
KV cache (fp16)368 KB per token

Source

Quantized bybartowski
FamilyGemma
Licensegemma
Downloads (HF)10,464
Last updated2024-08-03

Frequently asked questions about gemma-2-27b-it

How much VRAM does gemma-2-27b-it need?

About 20.2 GB at Q4_K_M with an 8K context window. That is the 16.6 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-2-27b-it on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 12.0 GB, so you would need a larger GPU or CPU offloading.

What is the file size of gemma-2-27b-it GGUF?

It depends on the quant: IQ2_XS is 8.4 GB, IQ2_S is 8.7 GB, IQ2_M is 9.4 GB, Q2_K is 10.4 GB, Q2_K_L is 10.7 GB, IQ3_XXS is 10.8 GB and 19 more.

What context length does gemma-2-27b-it support?

The GGUF header lists a trained context length of 8,192 tokens (8K). The KV cache grows with the context you use: about 3.1 GB at 8K and 3.1 GB at 8K.

Where can I download gemma-2-27b-it GGUF?

The files are on Hugging Face in the bartowski/gemma-2-27b-it-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-2-27b-it 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 Gemma‑2‑27B‑IT on 8 GB VRAM?

Use the Q4_K_S quant, which requires about 19.3 GB VRAM at 8‑k context. It will not fit on 8 GB VRAM without swapping.

Does this model support a system prompt?

No, the model does not support a system prompt.

What is the context length of this model?

The model supports a maximum context length of 8192 tokens.

Other GGUFs of gemma-2-27b-it

Similar models

Browse all →

More from bartowski

All by bartowski →

More Gemma models

All Gemma →

Other medium models

All medium models →