OP

Open Hermes 2.5 neural chat 3.1 InterlaceMerge 14b

OpenGGUF by S4schUpdated 2023-11-29
Download on Hugging Face
Parameters14.2B
Context32K
Layers64
Architecturellama
Smallest quant10.0 GB
Quick answer

Open Hermes 2.5 neural chat 3.1 InterlaceMerge 14b needs about 12.6 GB of memory at Q5_K_M (a 10.0 GB file) with an 8K context window, so it fits on a 16 GB GPU. This repository has 2 quants, from 10.0 GB (Q5_K_M) to 28.4 GB (F16). Longer context needs more: at 32K tokens the KV cache alone adds about 8.6 GB.

Open Hermes 2.5 neural chat 3.1 InterlaceMerge 14b is a 14.2B-parameter Open model in GGUF format, quantized by S4sch. Its trained context length is 32K tokens. The repository has 2 quant files, from 10.0 GB (Q5_K_M) to 28.4 GB (F16).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q5_K_M10.0 GB 12.6 GB
Too big Get file →
F1628.4 GB 31.1 GB
Too big Get file →

What GPU do you need for Open Hermes 2.5 neural chat 3.1 InterlaceMerge 14b?

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 Q5_K_M (10.0 GB file, about 12.6 GB in use).
  • 24 GB VRAM: the best quant that fits is Q5_K_M (10.0 GB file, about 12.6 GB in use).
  • 32 GB VRAM: the best quant that fits is F16 (28.4 GB file, about 31.1 GB in use).

Which Open Hermes 2.5 neural chat 3.1 InterlaceMerge 14b quant should you choose?

Q5_K_M (10.0 GB) is the middle option in this repository. Closer to the original model than 4-bit, for a moderate increase in size. A good pick when you have memory to spare.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturellama
Parameters14.2B
Layers64
Trained context32,768 tokens
KV cache (fp16)256 KB per token

Source

Quantized byS4sch
FamilyOpen
Licenseapache-2.0
Downloads (HF)127
Last updated2023-11-29

Frequently asked questions about Open Hermes 2.5 neural chat 3.1 InterlaceMerge 14b

How much VRAM does Open Hermes 2.5 neural chat 3.1 InterlaceMerge 14b need?

About 12.6 GB at Q5_K_M with an 8K context window. That is the 10.0 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 Open Hermes 2.5 neural chat 3.1 InterlaceMerge 14b on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 12.6 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Open Hermes 2.5 neural chat 3.1 InterlaceMerge 14b GGUF?

It depends on the quant: Q5_K_M is 10.0 GB, F16 is 28.4 GB.

What context length does Open Hermes 2.5 neural chat 3.1 InterlaceMerge 14b support?

The GGUF header lists a trained context length of 32,768 tokens (32K). The KV cache grows with the context you use: about 2.1 GB at 8K and 8.6 GB at 32K.

Where can I download Open Hermes 2.5 neural chat 3.1 InterlaceMerge 14b GGUF?

The files are on Hugging Face in the S4sch/Open-Hermes-2.5-neural-chat-3.1-InterlaceMerge-14b repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does Open Hermes 2.5 neural chat 3.1 InterlaceMerge 14b 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.

Similar models

Browse all →

More from S4sch

All by S4sch →

More Open models

All Open →

Other medium models

All medium models →