EE

EEVE Korean Instruct 10.8B v1.0 32k

EeveGGUF by williamjeong2Updated 2024-04-16
Download on Hugging Face
Parameters10.8B
Context32K
Layers48
Architecturellama
Smallest quant6.2 GB
Quick answer

EEVE Korean Instruct 10.8B v1.0 32k needs about 8.6 GB of memory at Q4_K_M (a 6.5 GB file) with an 8K context window, so it fits on a 12 GB GPU. This repository has 7 quants, from 6.2 GB (Q4_K_S) to 21.6 GB (F16). Longer context needs more: at 32K tokens the KV cache alone adds about 6.4 GB.

EEVE Korean Instruct 10.8B v1.0 32k is a 10.8B-parameter Eeve model in GGUF format, quantized by williamjeong2. Its trained context length is 32K tokens. The repository has 7 quant files, from 6.2 GB (Q4_K_S) to 21.6 GB (F16).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q4_K_S6.2 GB 8.3 GB
Too big Get file →
Q4_K_M6.5 GB 8.6 GB
Too big Get file →
Q5_K_S7.5 GB 9.6 GB
Too big Get file →
Q5_K_M7.7 GB 9.8 GB
Too big Get file →
Q6_K8.9 GB 11.0 GB
Too big Get file →
Q8_011.5 GB 13.6 GB
Too big Get file →
F1621.6 GB 23.7 GB
Too big Get file →

What GPU do you need for EEVE Korean Instruct 10.8B v1.0 32k?

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 Q6_K (8.9 GB file, about 11.0 GB in use).
  • 16 GB VRAM: the best quant that fits is Q8_0 (11.5 GB file, about 13.6 GB in use).
  • 24 GB VRAM: the best quant that fits is F16 (21.6 GB file, about 23.7 GB in use).

Which EEVE Korean Instruct 10.8B v1.0 32k quant should you choose?

Q4_K_M is the usual starting point for EEVE Korean Instruct 10.8B v1.0 32k: it balances file size and quality well. Here it is a 6.5 GB file.

The Q8_0 file is 11.5 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 Q4_K_S at 6.2 GB, with the quality trade-offs that come with fewer bits. The usual sweet spot between file size and quality, and the quant most people start with.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturellama
Parameters10.8B
Layers48
Trained context32,768 tokens
KV cache (fp16)192 KB per token

Source

Quantized bywilliamjeong2
FamilyEeve
Downloads (HF)135
Last updated2024-04-16

Frequently asked questions about EEVE Korean Instruct 10.8B v1.0 32k

How much VRAM does EEVE Korean Instruct 10.8B v1.0 32k need?

About 8.6 GB at Q4_K_M with an 8K context window. That is the 6.5 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 EEVE Korean Instruct 10.8B v1.0 32k on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 8.3 GB, so you would need a larger GPU or CPU offloading.

What is the file size of EEVE Korean Instruct 10.8B v1.0 32k GGUF?

It depends on the quant: Q4_K_S is 6.2 GB, Q4_K_M is 6.5 GB, Q5_K_S is 7.5 GB, Q5_K_M is 7.7 GB, Q6_K is 8.9 GB, Q8_0 is 11.5 GB and 1 more.

What context length does EEVE Korean Instruct 10.8B v1.0 32k support?

The GGUF header lists a trained context length of 32,768 tokens (32K). The KV cache grows with the context you use: about 1.6 GB at 8K and 6.4 GB at 32K.

Where can I download EEVE Korean Instruct 10.8B v1.0 32k GGUF?

The files are on Hugging Face in the williamjeong2/EEVE-Korean-Instruct-10.8B-v1.0-32k-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 williamjeong2

All by williamjeong2 →

More Eeve models

All Eeve →

Other medium models

All medium models →