PS

psyonic cetacean 20B iMat

PsyonicGGUF by InferenceIllusionistUpdated 2024-03-20
Download on Hugging Face
Parameters20B
Context4K
Layers62
Architecturellama
Smallest quant6.4 GB
Quick answer

psyonic cetacean 20B iMat needs about 19.1 GB of memory at IQ3_XS (a 8.2 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 7 quants, from 6.4 GB (IQ2_S) to 10.7 GB (IQ4_XS).

psyonic cetacean 20B iMat is a 20B-parameter Psyonic model in GGUF format, quantized by InferenceIllusionist. Its trained context length is 4K tokens. The repository has 7 quant files, from 6.4 GB (IQ2_S) to 10.7 GB (IQ4_XS).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ2_S6.4 GB 12.1 GB
Too big Get file →
IQ2_M6.9 GB 12.6 GB
Too big Get file →
IQ3_XXS7.6 GB 13.3 GB
Too big Get file →
IQ3_XS8.2 GB 13.9 GB
Too big Get file →
IQ3_S8.7 GB 14.4 GB
Too big Get file →
IQ3_M9.2 GB 14.9 GB
Too big Get file →
IQ4_XS10.7 GB 16.4 GB
Too big Get file →

What GPU do you need for psyonic cetacean 20B iMat?

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 IQ4_XS (10.7 GB file, about 21.6 GB in use).

Which psyonic cetacean 20B iMat quant should you choose?

IQ3_XS (8.2 GB) is the middle option in this repository. A reasonable choice when a 4-bit file will not fit, mostly on larger models. Quality loss is noticeable but often acceptable.

If memory is tight, the smallest file is IQ2_S at 6.4 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
Parameters20B
Layers62
Trained context4,096 tokens
KV cache (fp16)1240 KB per token

Source

FamilyPsyonic
Downloads (HF)3
Last updated2024-03-20

Frequently asked questions about psyonic cetacean 20B iMat

How much VRAM does psyonic cetacean 20B iMat need?

About 19.1 GB at IQ3_XS with an 8K context window. That is the 8.2 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 psyonic cetacean 20B iMat on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 17.3 GB, so you would need a larger GPU or CPU offloading.

What is the file size of psyonic cetacean 20B iMat GGUF?

It depends on the quant: IQ2_S is 6.4 GB, IQ2_M is 6.9 GB, IQ3_XXS is 7.6 GB, IQ3_XS is 8.2 GB, IQ3_S is 8.7 GB, IQ3_M is 9.2 GB and 1 more.

What context length does psyonic cetacean 20B iMat support?

The GGUF header lists a trained context length of 4,096 tokens (4K). The KV cache grows with the context you use: about 10.4 GB at 8K and 5.2 GB at 4K.

Where can I download psyonic cetacean 20B iMat GGUF?

The files are on Hugging Face in the InferenceIllusionist/psyonic-cetacean-20B-iMat-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 InferenceIllusionist

All by InferenceIllusionist →

More Psyonic models

All Psyonic →

Other medium models

All medium models →