XE

Xenon1 MetaModel moex8

XenonGGUF by RichardErkhovUpdated 2024-09-01
Download on Hugging Face
Parameters69.9B
Context4K
Layers—
Architecturellama
Smallest quant7.1 GB
Quick answer

Xenon1 MetaModel moex8 needs about 8.5 GB of memory at Q4_K_M (a 7.1 GB file) with an 8K context window, so it fits on a 12 GB GPU. This repository has 22 quants, from 7.1 GB (Q4_K_M) to 74.3 GB (Q8_0).

Xenon1 MetaModel moex8 is a 69.9B-parameter Xenon model in GGUF format, quantized by RichardErkhov. Its trained context length is 4K tokens. The repository has 22 quant files, from 7.1 GB (Q4_K_M) to 74.3 GB (Q8_0).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q4_K_M7.1 GB 8.0 GB
Too big Get file →
Q4_K12.6 GB 13.9 GB
Too big Get file →
Q5_014.0 GB 15.4 GB
Too big Get file →
IQ4_XS21.0 GB 22.8 GB
Too big Get file →
Q2_K25.9 GB 27.9 GB
Too big Get file →
IQ3_XS28.9 GB 31.2 GB
Too big Get file →
IQ3_S30.6 GB 32.9 GB
Too big Get file →
Q3_K_S30.6 GB 32.9 GB
Too big Get file →
IQ3_M32.1 GB 34.5 GB
Too big Get file →
Q3_K33.7 GB 36.3 GB
Too big Get file →
Q3_K_M33.7 GB 36.3 GB
Too big Get file →
Q4_134.9 GB 37.4 GB
Too big Get file →
Q3_K_L36.2 GB 38.8 GB
Too big Get file →
Q4_039.6 GB 42.4 GB
Too big Get file →
Q4_K_S39.8 GB (2 parts) 42.7 GB
Too big Get file →
IQ4_NL40.0 GB (2 parts) 42.9 GB
Too big Get file →
Q5_K_S48.2 GB (2 parts) 51.6 GB
Too big Get file →
Q5_K49.7 GB (2 parts) 53.2 GB
Too big Get file →
Q5_K_M49.7 GB (2 parts) 53.2 GB
Too big Get file →
Q5_152.6 GB (2 parts) 56.2 GB
Too big Get file →
Q6_K57.5 GB (2 parts) 61.4 GB
Too big Get file →
Q8_074.3 GB (2 parts) 79.3 GB
Too big Get file →

What GPU do you need for Xenon1 MetaModel moex8?

These figures add the file size, an fp16 KV cache at an 8K context window 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 Q4_K_M (7.1 GB file, about 8.5 GB in use).
  • 16 GB VRAM: the best quant that fits is Q4_K (12.6 GB file, about 14.6 GB in use).
  • 24 GB VRAM: the best quant that fits is IQ4_XS (21.0 GB file, about 24.0 GB in use).
  • 32 GB VRAM: the best quant that fits is Q2_K (25.9 GB file, about 29.5 GB in use).
  • 48 GB VRAM: the best quant that fits is IQ4_NL (40.0 GB file, about 45.3 GB in use).

Which Xenon1 MetaModel moex8 quant should you choose?

Q4_K_M is the usual starting point for Xenon1 MetaModel moex8: it balances file size and quality well. Here it is a 7.1 GB file.

The Q8_0 file is 74.3 GB, about 10.5 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.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturellama
Parameters69.9B
Trained context4,096 tokens

Source

Quantized byRichardErkhov
FamilyXenon
Downloads (HF)8
Last updated2024-09-01

Frequently asked questions about Xenon1 MetaModel moex8

How much VRAM does Xenon1 MetaModel moex8 need?

About 8.5 GB at Q4_K_M with an 8K context window. That is the 7.1 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 Xenon1 MetaModel moex8 on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 8.5 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Xenon1 MetaModel moex8 GGUF?

It depends on the quant: Q4_K_M is 7.1 GB, Q4_K is 12.6 GB, Q5_0 is 14.0 GB, IQ4_XS is 21.0 GB, Q2_K is 25.9 GB, IQ3_XS is 28.9 GB and 16 more.

What context length does Xenon1 MetaModel moex8 support?

The GGUF header lists a trained context length of 4,096 tokens (4K).

Where can I download Xenon1 MetaModel moex8 GGUF?

The files are on Hugging Face in the RichardErkhov/Xenon1_-_MetaModel_moex8-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 RichardErkhov

All by RichardErkhov →

More Xenon models

All Xenon →

Other large models

All large models →