SU

Sumatra 20b

SumatraGGUF by mradermacherUpdated 2026-10-22
Download on Hugging Face
Parameters21.5B
Context16K
Layers60
Architecturellama
Smallest quant8.1 GB
Quick answer

Sumatra 20b needs about 16.0 GB of memory at Q4_K_M (a 13.0 GB file) with an 8K context window, so it fits on a 16 GB GPU. This repository has 11 quants, from 8.1 GB (Q2_K) to 22.8 GB (Q8_0).

Sumatra 20b is a 21.5B-parameter Sumatra model in GGUF format, quantized by mradermacher. Its trained context length is 16K tokens. The repository has 11 quant files, from 8.1 GB (Q2_K) to 22.8 GB (Q8_0).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q2_K8.1 GB 11.1 GB
Too big Get file →
Q3_K_S9.4 GB 12.5 GB
Too big Get file →
Q3_K_M10.5 GB 13.5 GB
Too big Get file →
Q3_K_L11.4 GB 14.4 GB
Too big Get file →
IQ4_XS11.7 GB 14.7 GB
Too big Get file →
Q4_K_S12.3 GB 15.3 GB
Too big Get file →
Q4_K_M13.0 GB 16.0 GB
Too big Get file →
Q5_K_S14.8 GB 17.9 GB
Too big Get file →
Q5_K_M15.2 GB 18.2 GB
Too big Get file →
Q6_K17.6 GB 20.6 GB
Too big Get file →
Q8_022.8 GB 25.8 GB
Too big Get file →

What GPU do you need for Sumatra 20b?

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 Q2_K (8.1 GB file, about 11.1 GB in use).
  • 16 GB VRAM: the best quant that fits is Q4_K_M (13.0 GB file, about 16.0 GB in use).
  • 24 GB VRAM: the best quant that fits is Q6_K (17.6 GB file, about 20.6 GB in use).
  • 32 GB VRAM: the best quant that fits is Q8_0 (22.8 GB file, about 25.8 GB in use).

Which Sumatra 20b quant should you choose?

Q4_K_M is the usual starting point for Sumatra 20b: it balances file size and quality well. Here it is a 13.0 GB file.

The Q8_0 file is 22.8 GB, about 1.8 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 Q2_K at 8.1 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

Architecturellama
Parameters21.5B
Layers60
Trained context16,384 tokens
KV cache (fp16)300 KB per token

Source

Quantized bymradermacher
FamilySumatra
Downloads (HF)109
Last updated2026-10-22

Frequently asked questions about Sumatra 20b

How much VRAM does Sumatra 20b need?

About 16.0 GB at Q4_K_M with an 8K context window. That is the 13.0 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 Sumatra 20b on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 11.1 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Sumatra 20b GGUF?

It depends on the quant: Q2_K is 8.1 GB, Q3_K_S is 9.4 GB, Q3_K_M is 10.5 GB, Q3_K_L is 11.4 GB, IQ4_XS is 11.7 GB, Q4_K_S is 12.3 GB and 5 more.

What context length does Sumatra 20b support?

The GGUF header lists a trained context length of 16,384 tokens (16K). The KV cache grows with the context you use: about 2.5 GB at 8K and 5.0 GB at 16K.

Where can I download Sumatra 20b GGUF?

The files are on Hugging Face in the mradermacher/Sumatra-20b-GGUF 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 mradermacher

All by mradermacher →

More Sumatra models

All Sumatra →

Other medium models

All medium models →