GE

27b Kto

GemmaGGUF by anthracite-orgUpdated 2024-09-13
Download on Hugging Face
Parameters27.2B
Context8K
Layers46
Architecturegemma2
Smallest quant12.2 GB
Quick answer

27b Kto 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 7 quants, from 12.2 GB (IQ3_S) to 28.9 GB (Q8_0).

27b Kto is a 27.2B-parameter Gemma model in GGUF format, quantized by anthracite-org. Its trained context length is 8K tokens. The repository has 7 quant files, from 12.2 GB (IQ3_S) to 28.9 GB (Q8_0).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ3_S12.2 GB 15.8 GB
Too big Get file →
IQ3_M12.5 GB 16.0 GB
Too big Get file →
IQ4_XS14.8 GB 18.4 GB
Too big Get file →
Q4_K_M16.6 GB 20.2 GB
Too big Get file →
Q5_K_M19.4 GB 23.0 GB
Too big Get file →
Q6_K22.3 GB 25.9 GB
Too big Get file →
Q8_028.9 GB 32.5 GB
Too big Get file →

What GPU do you need for 27b Kto?

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: no quant fits at an 8K context.
  • 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_M (19.4 GB file, about 23.0 GB in use).
  • 32 GB VRAM: the best quant that fits is Q6_K (22.3 GB file, about 25.9 GB in use).
  • 48 GB VRAM: the best quant that fits is Q8_0 (28.9 GB file, about 32.5 GB in use).

Which 27b Kto quant should you choose?

Q4_K_M is the usual starting point for 27b Kto: 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 IQ3_S at 12.2 GB, with the quality trade-offs that come with fewer bits. A reasonable choice when a 4-bit file will not fit, mostly on larger models. Quality loss is noticeable but often acceptable.

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 byanthracite-org
FamilyGemma
Downloads (HF)72
Last updated2024-09-13

Frequently asked questions about 27b Kto

How much VRAM does 27b Kto 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 27b Kto on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 15.8 GB, so you would need a larger GPU or CPU offloading.

What is the file size of 27b Kto GGUF?

It depends on the quant: IQ3_S is 12.2 GB, IQ3_M is 12.5 GB, IQ4_XS is 14.8 GB, Q4_K_M is 16.6 GB, Q5_K_M is 19.4 GB, Q6_K is 22.3 GB and 1 more.

What context length does 27b Kto 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 27b Kto GGUF?

The files are on Hugging Face in the anthracite-org/magnum-v3-27b-kto-gguf repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

Other GGUFs of gemma-2-27b-chatml

Similar models

Browse all →

More from anthracite-org

All by anthracite-org →

More Gemma models

All Gemma →

Other medium models

All medium models →