MI

MistralNemoTiny

MistralGGUF by MaziyarPanahiUpdated 2024-10-31
Download on Hugging Face
Parameters4.61B
Context1000K
Layers12
Architecturellama
Smallest quant3.3 GB
Quick answer

MistralNemoTiny needs about 4.7 GB of memory at Q6_K (a 3.8 GB file) with an 8K context window, so it fits on a 6 GB GPU. This repository has 5 quants, from 3.3 GB (Q5_K_S) to 9.2 GB. Longer context needs more: at 32K tokens the KV cache alone adds about 1.6 GB.

MistralNemoTiny is a 4.61B-parameter Mistral model in GGUF format, quantized by MaziyarPanahi. Its trained context length is 1000K tokens. The repository has 5 quant files, from 3.3 GB (Q5_K_S) to 9.2 GB.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q5_K_S3.3 GB 4.2 GB
Fits Get file →
Q5_K_M3.3 GB 4.2 GB
Fits Get file →
Q6_K3.8 GB 4.7 GB
Fits Get file →
Q8_04.9 GB 5.8 GB
Fits Get file →
—9.2 GB 10.1 GB
Too big Get file →

What GPU do you need for MistralNemoTiny?

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 Q8_0 (4.9 GB file, about 5.8 GB in use).
  • 8 GB VRAM: the best quant that fits is Q8_0 (4.9 GB file, about 5.8 GB in use).
  • 12 GB VRAM: the best quant that fits is the main file (9.2 GB file, about 10.1 GB in use).

Which MistralNemoTiny quant should you choose?

Q6_K (3.8 GB) is the middle option in this repository. Very close to the original quality for most uses, at a clearly larger size than 4-bit.

If memory is tight, the smallest file is Q5_K_S at 3.3 GB, with the quality trade-offs that come with fewer bits. Closer to the original model than 4-bit, for a moderate increase in size. A good pick when you have memory to spare.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturellama
Parameters4.61B
Layers12
Trained context1,024,000 tokens
KV cache (fp16)48 KB per token

Source

Quantized byMaziyarPanahi
FamilyMistral
Downloads (HF)82
Last updated2024-10-31

Frequently asked questions about MistralNemoTiny

How much VRAM does MistralNemoTiny need?

About 4.7 GB at Q6_K with an 8K context window. That is the 3.8 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 MistralNemoTiny on an 8 GB GPU?

Yes, with Q8_0: the 4.9 GB file needs about 5.8 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of MistralNemoTiny GGUF?

It depends on the quant: Q5_K_S is 3.3 GB, Q5_K_M is 3.3 GB, Q6_K is 3.8 GB, Q8_0 is 4.9 GB, file is 9.2 GB.

What context length does MistralNemoTiny support?

The GGUF header lists a trained context length of 1,024,000 tokens (1000K). The KV cache grows with the context you use: about 0.4 GB at 8K and 1.6 GB at 32K.

Where can I download MistralNemoTiny GGUF?

The files are on Hugging Face in the MaziyarPanahi/MistralNemoTiny-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 MaziyarPanahi

All by MaziyarPanahi →

More Mistral models

All Mistral →

Other small models

All small models →