KR

krake v2 legacy

KrakeGGUF by NovelAIUpdated 2026-09-28
Download on Hugging Face
Parameters20.6B
Context2K
Layers44
Architecturegptneox
Smallest quant11.0 GB
Quick answer

krake v2 legacy needs about 26.2 GB of memory at Q6_K (a 16.9 GB file) with an 8K context window, so it fits on a 32 GB GPU. This repository has 3 quants, from 11.0 GB (IQ4_XS) to 21.9 GB (Q8_0).

Krake v2 Legacy – 20.6B GPT-NeoX finetune

Krake v2 Legacy is a 20.6B parameter GPT-NeoX model fine‑tuned by NovelAI. It was originally available to Opus subscribers and is now released under GPL‑3.0. The model supports a 2048 token context window.

Good at

  • Large language generation
  • Fine‑tuned text style
  • Contextual consistency

Limits

  • No vision support
  • GPL‑3.0 license restricts commercial use
  • Requires 8k context for memory estimates

Best for

  • Creative writing assistance
  • Chatbot development
  • Text summarization

Who it’s for. Developers and researchers needing a large, fine‑tuned GPT-NeoX model for text generation tasks. Suitable for those who can comply with GPL‑3.0 and have sufficient GPU memory.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ4_XS11.0 GB 13.7 GB
Too big Get file →
Q6_K16.9 GB 19.6 GB
Too big Get file →
Q8_021.9 GB 24.6 GB
Too big Get file →

For 8 GB VRAM, use IQ4_XS (11.02 GB file, ~20.4 GB memory at 8k context). For 12 GB VRAM, Q6_K (16.87 GB file, ~26.2 GB memory). For 16 GB VRAM, Q8_0 (21.85 GB file, ~31.2 GB memory). Memory estimates are approximate.

What GPU do you need for krake v2 legacy?

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: no quant fits at an 8K context.
  • 24 GB VRAM: the best quant that fits is IQ4_XS (11.0 GB file, about 20.4 GB in use).
  • 32 GB VRAM: the best quant that fits is Q8_0 (21.9 GB file, about 31.2 GB in use).

Which krake v2 legacy quant should you choose?

Q6_K (16.9 GB) is the middle option in this repository. Very close to the original quality for most uses, at a clearly larger size than 4-bit.

If memory is tight, the smallest file is IQ4_XS at 11.0 GB, with the quality trade-offs that come with fewer bits. The usual sweet spot between file size and quality, and the quant most people start with.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturegptneox
Parameters20.6B
Layers44
Trained context2,048 tokens
KV cache (fp16)1056 KB per token

Source

Quantized byNovelAI
FamilyKrake
Licensegpl-3.0
Downloads (HF)1,198
Last updated2026-09-28

Frequently asked questions about krake v2 legacy

How much VRAM does krake v2 legacy need?

About 26.2 GB at Q6_K with an 8K context window. That is the 16.9 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 krake v2 legacy on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 20.4 GB, so you would need a larger GPU or CPU offloading.

What is the file size of krake v2 legacy GGUF?

It depends on the quant: IQ4_XS is 11.0 GB, Q6_K is 16.9 GB, Q8_0 is 21.9 GB.

What context length does krake v2 legacy support?

The GGUF header lists a trained context length of 2,048 tokens (2K). The KV cache grows with the context you use: about 8.9 GB at 8K and 2.2 GB at 2K.

Where can I download krake v2 legacy GGUF?

The files are on Hugging Face in the NovelAI/krake-v2-legacy repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does krake v2 legacy use?

Hugging Face lists the license as gpl-3.0. 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 Krake v2 Legacy on 8 GB VRAM?

Yes, with the IQ4_XS quantized model, which requires about 20.4 GB of VRAM at 8k context, but actual VRAM usage may be lower with efficient batching.

Is Krake v2 Legacy suitable for vision tasks?

No, the model has no vision capabilities and is purely text‑based.

What license applies to Krake v2 Legacy?

The model is released under the GPL‑3.0 license, which allows modification but requires derivative works to remain open source.

Similar models

Browse all →

More from NovelAI

All by NovelAI →

Other medium models

All medium models →