NI

Ninja-V3

NinjaGGUF by MCZKUpdated 2024-07-03
Download on Hugging Face
Parameters7.24B
Context32K
Layers32
Architecturellama
Smallest quant2.8 GB
Quick answer

Ninja-V3 needs about 5.9 GB of memory at Q4_K_M (a 4.4 GB file) with an 8K context window, so it fits on a 6 GB GPU. This repository has 13 quants, from 2.8 GB (IQ3_XXS) to 14.5 GB (F16). Longer context needs more: at 32K tokens the KV cache alone adds about 4.3 GB.

Ninja-V3 is a 7.24B-parameter Ninja model in GGUF format, quantized by MCZK. Its trained context length is 32K tokens. The repository has 13 quant files, from 2.8 GB (IQ3_XXS) to 14.5 GB (F16).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ3_XXS2.8 GB 4.4 GB
Fits Get file →
IQ3_XS3.0 GB 4.6 GB
Fits Get file →
IQ3_S3.2 GB 4.8 GB
Fits Get file →
IQ3_M3.3 GB 4.9 GB
Fits Get file →
Q3_K_M3.5 GB 5.1 GB
Fits Get file →
IQ4_XS3.9 GB 5.5 GB
Fits Get file →
Q4_K_S4.1 GB 5.7 GB
Fits Get file →
Q4_K_M4.4 GB 5.9 GB
Fits Get file →
Q5_K_S5.0 GB 6.6 GB
Fits Get file →
Q5_K_M5.1 GB 6.7 GB
Fits Get file →
Q6_K5.9 GB 7.5 GB
Tight Get file →
Q8_07.7 GB 9.3 GB
Too big Get file →
F1614.5 GB 16.1 GB
Too big Get file →

What GPU do you need for Ninja-V3?

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: the best quant that fits is Q4_K_M (4.4 GB file, about 5.9 GB in use).
  • 8 GB VRAM: the best quant that fits is Q6_K (5.9 GB file, about 7.5 GB in use).
  • 12 GB VRAM: the best quant that fits is Q8_0 (7.7 GB file, about 9.3 GB in use).
  • 16 GB VRAM: the best quant that fits is Q8_0 (7.7 GB file, about 9.3 GB in use).
  • 24 GB VRAM: the best quant that fits is F16 (14.5 GB file, about 16.1 GB in use).

Which Ninja-V3 quant should you choose?

Q4_K_M is the usual starting point for Ninja-V3: it balances file size and quality well. Here it is a 4.4 GB file.

The Q8_0 file is 7.7 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 IQ3_XXS at 2.8 GB, with the quality trade-offs that come with fewer bits. A reasonable choice when a 4-bit file will not fit, mostly on larger models. Quality loss is noticeable but often acceptable.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturellama
Parameters7.24B
Layers32
Trained context32,768 tokens
KV cache (fp16)128 KB per token

Source

Quantized byMCZK
FamilyNinja
Downloads (HF)104
Last updated2024-07-03

Frequently asked questions about Ninja-V3

How much VRAM does Ninja-V3 need?

About 5.9 GB at Q4_K_M with an 8K context window. That is the 4.4 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 Ninja-V3 on an 8 GB GPU?

Yes, with Q6_K: the 5.9 GB file needs about 7.5 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of Ninja-V3 GGUF?

It depends on the quant: IQ3_XXS is 2.8 GB, IQ3_XS is 3.0 GB, IQ3_S is 3.2 GB, IQ3_M is 3.3 GB, Q3_K_M is 3.5 GB, IQ4_XS is 3.9 GB and 7 more.

What context length does Ninja-V3 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 Ninja-V3 GGUF?

The files are on Hugging Face in the MCZK/Ninja-V3-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

Other GGUFs of Ninja-V2-7B

Similar models

Browse all →

More from MCZK

All by MCZK →

More Ninja models

All Ninja →

Other small models

All small models →