NE

NeuralGuru

NeuralguruGGUF by mradermacherUpdated 2024-11-04
Download on Hugging Face
Parameters3.75B
Context32K
Layers32
Architecturellama
Smallest quant14.1 GB
Quick answer

NeuralGuru needs about 15.7 GB of memory at Q4_K_M (a 14.1 GB file) with an 8K context window, so it fits on a 16 GB GPU. This repository has 13 quants, from 14.1 GB (Q2_K) to 14.5 GB (F16). Longer context needs more: at 32K tokens the KV cache alone adds about 4.3 GB.

NeuralGuru is a 3.75B-parameter Neuralguru model in GGUF format, quantized by mradermacher. Its trained context length is 32K tokens. The repository has 13 quant files, from 14.1 GB (Q2_K) to 14.5 GB (F16).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q2_K14.1 GB 15.7 GB
Too big Get file →
Q3_K_L14.1 GB 15.7 GB
Too big Get file →
Q3_K_M14.1 GB 15.7 GB
Too big Get file →
Q3_K_S14.1 GB 15.7 GB
Too big Get file →
IQ4_XS14.1 GB 15.7 GB
Too big Get file →
Q4_014.1 GB 15.7 GB
Too big Get file →
Q4_K_M14.1 GB 15.7 GB
Too big Get file →
Q4_K_S14.1 GB 15.7 GB
Too big Get file →
Q5_K_M14.2 GB 15.7 GB
Too big Get file →
Q5_K_S14.2 GB 15.7 GB
Too big Get file →
Q6_K14.2 GB 15.7 GB
Too big Get file →
Q8_014.2 GB 15.8 GB
Too big Get file →
F1614.5 GB 16.1 GB
Too big Get file →

What GPU do you need for NeuralGuru?

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 Q8_0 (14.2 GB file, about 15.8 GB in use).
  • 24 GB VRAM: the best quant that fits is F16 (14.5 GB file, about 16.1 GB in use).

Which NeuralGuru quant should you choose?

Q4_K_M is the usual starting point for NeuralGuru: it balances file size and quality well. Here it is a 14.1 GB file.

The Q8_0 file is 14.2 GB, about 1.0 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 14.1 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
Parameters3.75B
Layers32
Trained context32,768 tokens
KV cache (fp16)128 KB per token

Source

Quantized bymradermacher
Licenseapache-2.0
Downloads (HF)173
Last updated2024-11-04

Frequently asked questions about NeuralGuru

How much VRAM does NeuralGuru need?

About 15.7 GB at Q4_K_M with an 8K context window. That is the 14.1 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 NeuralGuru on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 15.7 GB, so you would need a larger GPU or CPU offloading.

What is the file size of NeuralGuru GGUF?

It depends on the quant: Q2_K is 14.1 GB, Q3_K_L is 14.1 GB, Q3_K_M is 14.1 GB, Q3_K_S is 14.1 GB, IQ4_XS is 14.1 GB, Q4_0 is 14.1 GB and 7 more.

What context length does NeuralGuru support?

The GGUF header lists a trained context length of 32,768 tokens (32K). 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 NeuralGuru GGUF?

The files are on Hugging Face in the mradermacher/NeuralGuru-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 NeuralGuru 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 →

Other small models

All small models →