ER

OxytocinErosEngineering v0 4x7B passthrough

ErisGGUF by weezywitasneezyUpdated 2024-03-24
Download on Hugging Face
Parameters17.7B
Context32K
Layers80
Architecturellama
Smallest quant18.8 GB
Quick answer

OxytocinErosEngineering v0 4x7B passthrough needs about 22.0 GB of memory at Q8_0 (a 18.8 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 2 quants, from 18.8 GB (Q8_0) to 35.4 GB (F16). Longer context needs more: at 32K tokens the KV cache alone adds about 10.7 GB.

OxytocinErosEngineering v0 4x7B passthrough is a 17.7B-parameter Eris model in GGUF format, quantized by weezywitasneezy. Its trained context length is 32K tokens. The repository has 2 quant files, from 18.8 GB (Q8_0) to 35.4 GB (F16).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q8_018.8 GB 22.0 GB
Too big Get file →
F1635.4 GB 38.6 GB
Too big Get file →

What GPU do you need for OxytocinErosEngineering v0 4x7B passthrough?

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: no quant fits at an 8K context.
  • 24 GB VRAM: the best quant that fits is Q8_0 (18.8 GB file, about 22.0 GB in use).
  • 32 GB VRAM: the best quant that fits is Q8_0 (18.8 GB file, about 22.0 GB in use).
  • 48 GB VRAM: the best quant that fits is F16 (35.4 GB file, about 38.6 GB in use).

Which OxytocinErosEngineering v0 4x7B passthrough quant should you choose?

Q8_0 (18.8 GB) is the middle option in this repository. Close to lossless. It costs roughly twice the memory of a 4-bit file, so it suits cases where you have plenty of VRAM.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturellama
Parameters17.7B
Layers80
Trained context32,768 tokens
KV cache (fp16)320 KB per token

Source

Quantized byweezywitasneezy
FamilyEris
Downloads (HF)75
Last updated2024-03-24

Frequently asked questions about OxytocinErosEngineering v0 4x7B passthrough

How much VRAM does OxytocinErosEngineering v0 4x7B passthrough need?

About 22.0 GB at Q8_0 with an 8K context window. That is the 18.8 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 OxytocinErosEngineering v0 4x7B passthrough on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 22.0 GB, so you would need a larger GPU or CPU offloading.

What is the file size of OxytocinErosEngineering v0 4x7B passthrough GGUF?

It depends on the quant: Q8_0 is 18.8 GB, F16 is 35.4 GB.

What context length does OxytocinErosEngineering v0 4x7B passthrough support?

The GGUF header lists a trained context length of 32,768 tokens (32K). The KV cache grows with the context you use: about 2.7 GB at 8K and 10.7 GB at 32K.

Where can I download OxytocinErosEngineering v0 4x7B passthrough GGUF?

The files are on Hugging Face in the weezywitasneezy/OxytocinErosEngineering_v0-4x7B-passthrough repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

Other GGUFs of Eris_Remix_7B

Similar models

Browse all →

More from weezywitasneezy

All by weezywitasneezy →

More Eris models

All Eris →

Other medium models

All medium models →