ME

Merged Vicuna RP Stew 34B

MergedGGUF by tachyphylaxisUpdated 2024-03-03
Download on Hugging Face
Parameters34.4B
Context32K
Layers60
Architecturellama
Smallest quant19.7 GB
Quick answer

Merged Vicuna RP Stew 34B needs about 23.2 GB of memory at Q4_K_M (a 20.7 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 5 quants, from 19.7 GB (IQ4_NL) to 36.5 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 8.1 GB.

Merged Vicuna RP Stew 34B is a 34.4B-parameter Merged model in GGUF format, quantized by tachyphylaxis. Its trained context length is 32K tokens. The repository has 5 quant files, from 19.7 GB (IQ4_NL) to 36.5 GB (Q8_0).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ4_NL19.7 GB 22.2 GB
Too big Get file →
Q4_K_M20.7 GB 23.2 GB
Too big Get file →
Q5_K_M24.3 GB 26.8 GB
Too big Get file →
Q6_K28.2 GB 30.7 GB
Too big Get file →
Q8_036.5 GB 39.1 GB
Too big Get file →

What GPU do you need for Merged Vicuna RP Stew 34B?

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 Q4_K_M (20.7 GB file, about 23.2 GB in use).
  • 32 GB VRAM: the best quant that fits is Q6_K (28.2 GB file, about 30.7 GB in use).
  • 48 GB VRAM: the best quant that fits is Q8_0 (36.5 GB file, about 39.1 GB in use).

Which Merged Vicuna RP Stew 34B quant should you choose?

Q4_K_M is the usual starting point for Merged Vicuna RP Stew 34B: it balances file size and quality well. Here it is a 20.7 GB file.

The Q8_0 file is 36.5 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 IQ4_NL at 19.7 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
Parameters34.4B
Layers60
Trained context32,768 tokens
KV cache (fp16)240 KB per token

Source

Quantized bytachyphylaxis
FamilyMerged
Downloads (HF)77
Last updated2024-03-03

Frequently asked questions about Merged Vicuna RP Stew 34B

How much VRAM does Merged Vicuna RP Stew 34B need?

About 23.2 GB at Q4_K_M with an 8K context window. That is the 20.7 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 Merged Vicuna RP Stew 34B on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 22.2 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Merged Vicuna RP Stew 34B GGUF?

It depends on the quant: IQ4_NL is 19.7 GB, Q4_K_M is 20.7 GB, Q5_K_M is 24.3 GB, Q6_K is 28.2 GB, Q8_0 is 36.5 GB.

What context length does Merged Vicuna RP Stew 34B support?

The GGUF header lists a trained context length of 32,768 tokens (32K). The KV cache grows with the context you use: about 2.0 GB at 8K and 8.1 GB at 32K.

Where can I download Merged Vicuna RP Stew 34B GGUF?

The files are on Hugging Face in the tachyphylaxis/Merged-Vicuna-RP-Stew-34B-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 tachyphylaxis

All by tachyphylaxis →

More Merged models

All Merged →

Other large models

All large models →