UM

Umbra MoE 4x10.7

UmbraGGUF by mradermacherUpdated 2024-12-02
Download on Hugging Face
Parameters36.1B
Context4K
Layers48
Architecturellama
Smallest quant13.2 GB
Quick answer

Umbra MoE 4x10.7 needs about 23.9 GB of memory at Q4_K_M (a 21.8 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 11 quants, from 13.2 GB (Q2_K) to 38.4 GB (Q8_0).

Umbra MoE 4x10.7 is a 36.1B-parameter Umbra model in GGUF format, quantized by mradermacher. Its trained context length is 4K tokens. The repository has 11 quant files, from 13.2 GB (Q2_K) to 38.4 GB (Q8_0).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q2_K13.2 GB 14.5 GB
Too big Get file →
Q3_K_S15.6 GB 16.9 GB
Too big Get file →
Q3_K_M17.3 GB 18.6 GB
Too big Get file →
Q3_K_L18.7 GB 20.0 GB
Too big Get file →
IQ4_XS19.5 GB 20.8 GB
Too big Get file →
Q4_K_S20.5 GB 21.8 GB
Too big Get file →
Q4_K_M21.8 GB 23.1 GB
Too big Get file →
Q5_K_S24.8 GB 26.1 GB
Too big Get file →
Q5_K_M25.6 GB 26.9 GB
Too big Get file →
Q6_K29.6 GB 30.9 GB
Too big Get file →
Q8_038.4 GB 39.7 GB
Too big Get file →

What GPU do you need for Umbra MoE 4x10.7?

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 Q2_K (13.2 GB file, about 15.3 GB in use).
  • 24 GB VRAM: the best quant that fits is Q4_K_M (21.8 GB file, about 23.9 GB in use).
  • 32 GB VRAM: the best quant that fits is Q6_K (29.6 GB file, about 31.7 GB in use).
  • 48 GB VRAM: the best quant that fits is Q8_0 (38.4 GB file, about 40.5 GB in use).

Which Umbra MoE 4x10.7 quant should you choose?

Q4_K_M is the usual starting point for Umbra MoE 4x10.7: it balances file size and quality well. Here it is a 21.8 GB file.

The Q8_0 file is 38.4 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 13.2 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
Parameters36.1B (approximate, mixture of experts)
Layers48
Trained context4,096 tokens
KV cache (fp16)192 KB per token

Source

Quantized bymradermacher
FamilyUmbra
Licenseapache-2.0
Downloads (HF)184
Last updated2024-12-02

Frequently asked questions about Umbra MoE 4x10.7

How much VRAM does Umbra MoE 4x10.7 need?

About 23.9 GB at Q4_K_M with an 8K context window. That is the 21.8 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 Umbra MoE 4x10.7 on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 15.3 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Umbra MoE 4x10.7 GGUF?

It depends on the quant: Q2_K is 13.2 GB, Q3_K_S is 15.6 GB, Q3_K_M is 17.3 GB, Q3_K_L is 18.7 GB, IQ4_XS is 19.5 GB, Q4_K_S is 20.5 GB and 5 more.

What context length does Umbra MoE 4x10.7 support?

The GGUF header lists a trained context length of 4,096 tokens (4K). The KV cache grows with the context you use: about 1.6 GB at 8K and 0.8 GB at 4K.

Where can I download Umbra MoE 4x10.7 GGUF?

The files are on Hugging Face in the mradermacher/Umbra-MoE-4x10.7-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 Umbra MoE 4x10.7 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.

Other GGUFs of Umbra-MoE-4x10.7

Similar models

Browse all →

More from mradermacher

All by mradermacher →

More Umbra models

All Umbra →

Other large models

All large models →