MI

Mistral Small_ChatML

MistralGGUF by Sicarius-PrototypingUpdated 2024-09-25
Download on Hugging Face
Parameters22.2B
Context128K
Layers56
Architecturellama
Smallest quant9.6 GB
Quick answer

Mistral Small_ChatML needs about 15.7 GB of memory at Q4_K_M (a 13.3 GB file) with an 8K context window, so it fits on a 16 GB GPU. This repository has 9 quants, from 9.6 GB (Q3_K_S) to 23.6 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 7.5 GB.

Mistral Small_ChatML is a 22.2B-parameter Mistral model in GGUF format, quantized by Sicarius-Prototyping. Its trained context length is 128K tokens. The repository has 9 quant files, from 9.6 GB (Q3_K_S) to 23.6 GB (Q8_0).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q3_K_S9.6 GB 12.0 GB
Too big Get file →
Q3_K_M10.8 GB 13.1 GB
Too big Get file →
Q3_K_L11.7 GB 14.1 GB
Too big Get file →
Q4_K_S12.7 GB 15.0 GB
Too big Get file →
Q4_K_M13.3 GB 15.7 GB
Too big Get file →
Q5_K_S15.3 GB 17.7 GB
Too big Get file →
Q5_K_M15.7 GB 18.1 GB
Too big Get file →
Q6_K18.3 GB 20.6 GB
Too big Get file →
Q8_023.6 GB 26.0 GB
Too big Get file →

What GPU do you need for Mistral Small_ChatML?

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: the best quant that fits is Q3_K_S (9.6 GB file, about 12.0 GB in use).
  • 16 GB VRAM: the best quant that fits is Q4_K_M (13.3 GB file, about 15.7 GB in use).
  • 24 GB VRAM: the best quant that fits is Q6_K (18.3 GB file, about 20.6 GB in use).
  • 32 GB VRAM: the best quant that fits is Q8_0 (23.6 GB file, about 26.0 GB in use).

Which Mistral Small_ChatML quant should you choose?

Q4_K_M is the usual starting point for Mistral Small_ChatML: it balances file size and quality well. Here it is a 13.3 GB file.

The Q8_0 file is 23.6 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 Q3_K_S at 9.6 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
Parameters22.2B
Layers56
Trained context131,072 tokens
KV cache (fp16)224 KB per token

Source

FamilyMistral
Licenseapache-2.0
Downloads (HF)122
Last updated2024-09-25

Frequently asked questions about Mistral Small_ChatML

How much VRAM does Mistral Small_ChatML need?

About 15.7 GB at Q4_K_M with an 8K context window. That is the 13.3 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 Mistral Small_ChatML on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 12.0 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Mistral Small_ChatML GGUF?

It depends on the quant: Q3_K_S is 9.6 GB, Q3_K_M is 10.8 GB, Q3_K_L is 11.7 GB, Q4_K_S is 12.7 GB, Q4_K_M is 13.3 GB, Q5_K_S is 15.3 GB and 3 more.

What context length does Mistral Small_ChatML support?

The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 1.9 GB at 8K and 7.5 GB at 32K.

Where can I download Mistral Small_ChatML GGUF?

The files are on Hugging Face in the Sicarius-Prototyping/Mistral-Small_ChatML_GGUFs repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does Mistral Small_ChatML 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 Sicarius-Prototyping

All by Sicarius-Prototyping →

More Mistral models

All Mistral →

Other medium models

All medium models →