NE

NEBULA 23B v1.0

NebulaGGUF by mradermacherUpdated 2024-12-16
Download on Hugging Face
Parameters23.8B
Context4K
Layers108
Architecturellama
Smallest quant8.8 GB
Quick answer

NEBULA 23B v1.0 needs about 18.4 GB of memory at Q4_K_M (a 14.3 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 14 quants, from 8.8 GB (Q2_K) to 25.3 GB (Q8_0).

NEBULA 23B v1.0 is a 23.8B-parameter Nebula model in GGUF format, quantized by mradermacher. Its trained context length is 4K tokens. The repository has 14 quant files, from 8.8 GB (Q2_K) to 25.3 GB (Q8_0).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q2_K8.8 GB 11.1 GB
Too big Get file →
IQ3_XS9.8 GB 12.1 GB
Too big Get file →
Q3_K_S10.3 GB 12.6 GB
Too big Get file →
IQ3_S10.3 GB 12.7 GB
Too big Get file →
IQ3_M10.7 GB 13.0 GB
Too big Get file →
Q3_K_M11.5 GB 13.8 GB
Too big Get file →
Q3_K_L12.5 GB 14.8 GB
Too big Get file →
IQ4_XS12.9 GB 15.2 GB
Too big Get file →
Q4_K_S13.5 GB 15.8 GB
Too big Get file →
Q4_K_M14.3 GB 16.6 GB
Too big Get file →
Q5_K_S16.4 GB 18.7 GB
Too big Get file →
Q5_K_M16.8 GB 19.2 GB
Too big Get file →
Q6_K19.5 GB 21.9 GB
Too big Get file →
Q8_025.3 GB 27.6 GB
Too big Get file →

What GPU do you need for NEBULA 23B v1.0?

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 (11.5 GB file, about 15.6 GB in use).
  • 24 GB VRAM: the best quant that fits is Q6_K (19.5 GB file, about 23.7 GB in use).
  • 32 GB VRAM: the best quant that fits is Q8_0 (25.3 GB file, about 29.4 GB in use).

Which NEBULA 23B v1.0 quant should you choose?

Q4_K_M is the usual starting point for NEBULA 23B v1.0: it balances file size and quality well. Here it is a 14.3 GB file.

The Q8_0 file is 25.3 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 8.8 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
Parameters23.8B
Layers108
Trained context4,096 tokens
KV cache (fp16)432 KB per token

Source

Quantized bymradermacher
FamilyNebula
Licenseapache-2.0
Downloads (HF)155
Last updated2024-12-16

Frequently asked questions about NEBULA 23B v1.0

How much VRAM does NEBULA 23B v1.0 need?

About 18.4 GB at Q4_K_M with an 8K context window. That is the 14.3 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 NEBULA 23B v1.0 on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 12.9 GB, so you would need a larger GPU or CPU offloading.

What is the file size of NEBULA 23B v1.0 GGUF?

It depends on the quant: Q2_K is 8.8 GB, IQ3_XS is 9.8 GB, Q3_K_S is 10.3 GB, IQ3_S is 10.3 GB, IQ3_M is 10.7 GB, Q3_K_M is 11.5 GB and 8 more.

What context length does NEBULA 23B v1.0 support?

The GGUF header lists a trained context length of 4,096 tokens (4K). The KV cache grows with the context you use: about 3.6 GB at 8K and 1.8 GB at 4K.

Where can I download NEBULA 23B v1.0 GGUF?

The files are on Hugging Face in the mradermacher/NEBULA-23B-v1.0-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 NEBULA 23B v1.0 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.

Other GGUFs of NEBULA-23.8B-v1.0

Similar models

Browse all →

More from mradermacher

All by mradermacher →

More Nebula models

All Nebula →

Other medium models

All medium models →