WE

WeirdSlerp 30B

WeirdslerpGGUF by mradermacherUpdated 2024-10-22
Download on Hugging Face
Parameters29.5B
Context128K
Layers32
Architecturellama
Smallest quant11.2 GB
Quick answer

WeirdSlerp 30B needs about 19.4 GB of memory at Q4_K_M (a 17.9 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 11 quants, from 11.2 GB (Q2_K) to 31.3 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 4.3 GB.

WeirdSlerp 30B is a 29.5B-parameter Weirdslerp model in GGUF format, quantized by mradermacher. Its trained context length is 128K tokens. The repository has 11 quant files, from 11.2 GB (Q2_K) to 31.3 GB (Q8_0).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q2_K11.2 GB 12.8 GB
Too big Get file →
Q3_K_S13.1 GB 14.7 GB
Too big Get file →
Q3_K_M14.5 GB 16.0 GB
Too big Get file →
Q3_K_L15.6 GB 17.2 GB
Too big Get file →
IQ4_XS16.2 GB 17.7 GB
Too big Get file →
Q4_K_S17.0 GB 18.6 GB
Too big Get file →
Q4_K_M17.9 GB 19.4 GB
Too big Get file →
Q5_K_S20.4 GB 22.0 GB
Too big Get file →
Q5_K_M20.9 GB 22.5 GB
Too big Get file →
Q6_K24.2 GB 25.8 GB
Too big Get file →
Q8_031.3 GB 32.9 GB
Too big Get file →

What GPU do you need for WeirdSlerp 30B?

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 Q3_K_M (14.5 GB file, about 16.0 GB in use).
  • 24 GB VRAM: the best quant that fits is Q5_K_M (20.9 GB file, about 22.5 GB in use).
  • 32 GB VRAM: the best quant that fits is Q6_K (24.2 GB file, about 25.8 GB in use).
  • 48 GB VRAM: the best quant that fits is Q8_0 (31.3 GB file, about 32.9 GB in use).

Which WeirdSlerp 30B quant should you choose?

Q4_K_M is the usual starting point for WeirdSlerp 30B: it balances file size and quality well. Here it is a 17.9 GB file.

The Q8_0 file is 31.3 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 Q2_K at 11.2 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
Parameters29.5B
Layers32
Trained context131,072 tokens
KV cache (fp16)128 KB per token

Source

Quantized bymradermacher
Licenseapache-2.0
Downloads (HF)3
Last updated2024-10-22

Frequently asked questions about WeirdSlerp 30B

How much VRAM does WeirdSlerp 30B need?

About 19.4 GB at Q4_K_M with an 8K context window. That is the 17.9 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 WeirdSlerp 30B on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 12.8 GB, so you would need a larger GPU or CPU offloading.

What is the file size of WeirdSlerp 30B GGUF?

It depends on the quant: Q2_K is 11.2 GB, Q3_K_S is 13.1 GB, Q3_K_M is 14.5 GB, Q3_K_L is 15.6 GB, IQ4_XS is 16.2 GB, Q4_K_S is 17.0 GB and 5 more.

What context length does WeirdSlerp 30B support?

The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 1.1 GB at 8K and 4.3 GB at 32K.

Where can I download WeirdSlerp 30B GGUF?

The files are on Hugging Face in the mradermacher/WeirdSlerp-30B-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does WeirdSlerp 30B 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 mradermacher

All by mradermacher →

More Weirdslerp models

All Weirdslerp →

Other medium models

All medium models →