TH

K2 Think V2

ThinkGGUF by benjaminradioUpdated 2026-05-13
Download on Hugging Face
Parameters72.6B
Context256K
Layers80
Architecturellama
Smallest quant59.5 GB
Quick answer

K2 Think V2 needs about 62.7 GB of memory at Q6_K (a 59.5 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 2 quants, from 59.5 GB (Q6_K) to 290.2 GB (F32). Longer context needs more: at 32K tokens the KV cache alone adds about 10.7 GB.

K2 Think V2 is a 72.6B-parameter Think model in GGUF format, quantized by benjaminradio. Its trained context length is 256K tokens. The repository has 2 quant files, from 59.5 GB (Q6_K) to 290.2 GB (F32).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q6_K59.5 GB (4 parts) 62.7 GB
Too big Get file →
F32290.2 GB (16 parts) 293.4 GB
Too big Get file →

What GPU do you need for K2 Think V2?

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: no quant fits at an 8K context.
  • 32 GB VRAM: no quant fits at an 8K context.
  • 48 GB VRAM: no quant fits at an 8K context.

Which K2 Think V2 quant should you choose?

Q6_K (59.5 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.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturellama
Parameters72.6B
Layers80
Trained context262,144 tokens
KV cache (fp16)320 KB per token

Source

Quantized bybenjaminradio
FamilyThink
Licenseapache-2.0
Downloads (HF)45
Last updated2026-05-13

Frequently asked questions about K2 Think V2

How much VRAM does K2 Think V2 need?

About 62.7 GB at Q6_K with an 8K context window. That is the 59.5 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 K2 Think V2 on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 62.7 GB, so you would need a larger GPU or CPU offloading.

What is the file size of K2 Think V2 GGUF?

It depends on the quant: Q6_K is 59.5 GB, F32 is 290.2 GB.

What context length does K2 Think V2 support?

The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 2.7 GB at 8K and 10.7 GB at 32K.

Where can I download K2 Think V2 GGUF?

The files are on Hugging Face in the benjaminradio/K2-Think-V2-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 K2 Think V2 use?

Hugging Face lists the license as apache-2.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.

Similar models

Browse all →

More Think models

All Think →

Other large models

All large models →