PS

PsyMedRP XL v2 i1

PsymedrpGGUF by mradermacherUpdated 2024-09-27
Download on Hugging Face
Parameters27.6B
Context4K
Layers86
Architecturellama
Smallest quant6.0 GB
Quick answer

PsyMedRP XL v2 i1 needs about 31.6 GB of memory at Q4_K_M (a 16.6 GB file) with an 8K context window, so it fits on a 32 GB GPU. This repository has 21 quants, from 6.0 GB (IQ1_S) to 22.7 GB (Q6_K).

PsyMedRP XL v2 i1 is a 27.6B-parameter Psymedrp model in GGUF format, quantized by mradermacher. Its trained context length is 4K tokens. The repository has 21 quant files, from 6.0 GB (IQ1_S) to 22.7 GB (Q6_K).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ1_S6.0 GB 13.7 GB
Too big Get file →
IQ1_M6.5 GB 14.3 GB
Too big Get file →
IQ2_XXS7.4 GB 15.1 GB
Too big Get file →
IQ2_XS8.2 GB 15.9 GB
Too big Get file →
IQ2_S8.8 GB 16.5 GB
Too big Get file →
IQ2_M9.5 GB 17.2 GB
Too big Get file →
Q2_K10.2 GB 17.9 GB
Too big Get file →
IQ3_XXS10.5 GB 18.2 GB
Too big Get file →
IQ3_XS11.3 GB 19.0 GB
Too big Get file →
IQ3_S11.9 GB 19.6 GB
Too big Get file →
Q3_K_S11.9 GB 19.6 GB
Too big Get file →
IQ3_M12.6 GB 20.3 GB
Too big Get file →
Q3_K_M13.4 GB 21.1 GB
Too big Get file →
Q3_K_L14.7 GB 22.4 GB
Too big Get file →
IQ4_XS14.7 GB 22.4 GB
Too big Get file →
Q4_015.6 GB 23.3 GB
Too big Get file →
Q4_K_S15.7 GB 23.4 GB
Too big Get file →
Q4_K_M16.6 GB 24.3 GB
Too big Get file →
Q5_K_S19.0 GB 26.7 GB
Too big Get file →
Q5_K_M19.5 GB 27.3 GB
Too big Get file →
Q6_K22.7 GB 30.4 GB
Too big Get file →

What GPU do you need for PsyMedRP XL v2 i1?

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 IQ2_S (8.8 GB file, about 23.7 GB in use).
  • 32 GB VRAM: the best quant that fits is Q4_K_M (16.6 GB file, about 31.6 GB in use).
  • 48 GB VRAM: the best quant that fits is Q6_K (22.7 GB file, about 37.6 GB in use).

Which PsyMedRP XL v2 i1 quant should you choose?

Q4_K_M is the usual starting point for PsyMedRP XL v2 i1: it balances file size and quality well. Here it is a 16.6 GB file.

If memory is tight, the smallest file is IQ1_S at 6.0 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
Parameters27.6B
Layers86
Trained context4,096 tokens
KV cache (fp16)1720 KB per token

Source

Quantized bymradermacher
FamilyPsymedrp
Downloads (HF)5
Last updated2024-09-27

Frequently asked questions about PsyMedRP XL v2 i1

How much VRAM does PsyMedRP XL v2 i1 need?

About 31.6 GB at Q4_K_M with an 8K context window. That is the 16.6 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 PsyMedRP XL v2 i1 on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 21.0 GB, so you would need a larger GPU or CPU offloading.

What is the file size of PsyMedRP XL v2 i1 GGUF?

It depends on the quant: IQ1_S is 6.0 GB, IQ1_M is 6.5 GB, IQ2_XXS is 7.4 GB, IQ2_XS is 8.2 GB, IQ2_S is 8.8 GB, IQ2_M is 9.5 GB and 15 more.

What context length does PsyMedRP XL v2 i1 support?

The GGUF header lists a trained context length of 4,096 tokens (4K). The KV cache grows with the context you use: about 14.4 GB at 8K and 7.2 GB at 4K.

Where can I download PsyMedRP XL v2 i1 GGUF?

The files are on Hugging Face in the mradermacher/PsyMedRP_XL_v2-i1-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 mradermacher

All by mradermacher →

More Psymedrp models

All Psymedrp →

Other medium models

All medium models →