SA

San x2 t2 Q6 K M

SanGGUF by bruhzairUpdated 2025-03-01
Download on Hugging Face
Parameters91.1B
Context128K
Layers104
Architecturellama
Smallest quant74.7 GB
Quick answer

San x2 t2 Q6 K M needs about 78.7 GB of memory at Q6_K (a 74.7 GB file) with an 8K context window, which is more than a 48 GB GPU holds. Longer context needs more: at 32K tokens the KV cache alone adds about 14.0 GB.

San x2 t2 Q6 K M is a 91.1B-parameter San model in GGUF format, quantized by bruhzair. Its trained context length is 128K tokens. The repository has 1 quant file, from 74.7 GB (Q6_K) to 74.7 GB (Q6_K).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q6_K74.7 GB (2 parts) 78.7 GB
Too big Get file →

What GPU do you need for San x2 t2 Q6 K M?

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 San x2 t2 Q6 K M quant should you choose?

Q6_K (74.7 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
Parameters91.1B
Layers104
Trained context131,072 tokens
KV cache (fp16)416 KB per token

Source

Quantized bybruhzair
FamilySan
Downloads (HF)6
Last updated2025-03-01

Frequently asked questions about San x2 t2 Q6 K M

How much VRAM does San x2 t2 Q6 K M need?

About 78.7 GB at Q6_K with an 8K context window. That is the 74.7 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 San x2 t2 Q6 K M on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 78.7 GB, so you would need a larger GPU or CPU offloading.

What is the file size of San x2 t2 Q6 K M GGUF?

It depends on the quant: Q6_K is 74.7 GB.

What context length does San x2 t2 Q6 K M support?

The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 3.5 GB at 8K and 14.0 GB at 32K.

Where can I download San x2 t2 Q6 K M GGUF?

The files are on Hugging Face in the bruhzair/San-x2-t2-Q6_K_M repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

Similar models

Browse all →

More from bruhzair

All by bruhzair →

More San models

All San →

Other large models

All large models →