You can deploy this model and start earning money today!

bartowski/
Q2.5-MS-Mistoria-72b-v2-GGUF

Name: Q2.5-MS-Mistoria-72b-v2-GGUF
Availability: InStock

quantized

SteelStorage/Q2.5-MS-Mistoria-72b-v2

Views

No views yet

Llamacpp imatrix Quantizations of Q2.5-MS-Mistoria-72b-v2

Using llama.cpp release b4132 for quantization.

Original model: https://huggingface.co/Steelskull/Q2.5-MS-Mistoria-72b-v2

All quants made using imatrix option with dataset from here

Run them in LM Studio

Prompt format

<|im_start|>system
{system_prompt}<|im_end|>
<|im_start|>user
{prompt}<|im_end|>
<|im_start|>assistant

Download a file (not the whole branch) from below:

Filename	Quant type	File Size	Split	Description
Q2.5-MS-Mistoria-72b-v2-Q8_0.gguf	Q8_0	77.26GB	true	Extremely high quality, generally unneeded but max available quant.
Q2.5-MS-Mistoria-72b-v2-Q6_K.gguf	Q6_K	64.35GB	true	Very high quality, near perfect, recommended.
Q2.5-MS-Mistoria-72b-v2-Q5_K_M.gguf	Q5_K_M	54.45GB	true	High quality, recommended.
Q2.5-MS-Mistoria-72b-v2-Q5_K_S.gguf	Q5_K_S	51.38GB	true	High quality, recommended.
Q2.5-MS-Mistoria-72b-v2-Q4_K_M.gguf	Q4_K_M	47.42GB	false	Good quality, default size for most use cases, recommended.
Q2.5-MS-Mistoria-72b-v2-Q4_K_S.gguf	Q4_K_S	43.89GB	false	Slightly lower quality with more space savings, recommended.
Q2.5-MS-Mistoria-72b-v2-Q4_0.gguf	Q4_0	41.38GB	false	Legacy format, generally not worth using over similarly sized formats
Q2.5-MS-Mistoria-72b-v2-Q4_0_8_8.gguf	Q4_0_8_8	41.23GB	false	Optimized for ARM and AVX inference. Requires 'sve' support for ARM (see details below). Don't use on Mac.
Q2.5-MS-Mistoria-72b-v2-Q3_K_XL.gguf	Q3_K_XL	40.60GB	false	Uses Q8_0 for embed and output weights. Lower quality but usable, good for low RAM availability.
Q2.5-MS-Mistoria-72b-v2-IQ4_XS.gguf	IQ4_XS	39.71GB	false	Decent quality, smaller than Q4_K_S with similar performance, recommended.
Q2.5-MS-Mistoria-72b-v2-Q3_K_L.gguf	Q3_K_L	39.51GB	false	Lower quality but usable, good for low RAM availability.
Q2.5-MS-Mistoria-72b-v2-Q3_K_M.gguf	Q3_K_M	37.70GB	false	Low quality.
Q2.5-MS-Mistoria-72b-v2-IQ3_M.gguf	IQ3_M	35.50GB	false	Medium-low quality, new method with decent performance comparable to Q3_K_M.
Q2.5-MS-Mistoria-72b-v2-Q3_K_S.gguf	Q3_K_S	34.49GB	false	Low quality, not recommended.
Q2.5-MS-Mistoria-72b-v2-IQ3_XXS.gguf	IQ3_XXS	31.85GB	false	Lower quality, new method with decent performance, comparable to Q3 quants.
Q2.5-MS-Mistoria-72b-v2-Q2_K_L.gguf	Q2_K_L	31.03GB	false	Uses Q8_0 for embed and output weights. Very low quality but surprisingly usable.
Q2.5-MS-Mistoria-72b-v2-Q2_K.gguf	Q2_K	29.81GB	false	Very low quality but surprisingly usable.
Q2.5-MS-Mistoria-72b-v2-IQ2_M.gguf	IQ2_M	29.34GB	false	Relatively low quality, uses SOTA techniques to be surprisingly usable.
Q2.5-MS-Mistoria-72b-v2-IQ2_XS.gguf	IQ2_XS	27.06GB	false	Low quality, uses SOTA techniques to be usable.
Q2.5-MS-Mistoria-72b-v2-IQ2_XXS.gguf	IQ2_XXS	25.49GB	false	Very low quality, uses SOTA techniques to be usable.
Q2.5-MS-Mistoria-72b-v2-IQ1_M.gguf	IQ1_M	23.74GB	false	Extremely low quality, not recommended.

Embed/output weights

Some of these quants (Q3_K_XL, Q4_K_L etc) are the standard quantization method with the embeddings and output weights quantized to Q8_0 instead of what they would normally default to.

Downloading using huggingface-cli

Click to view download instructions

First, make sure you have hugginface-cli installed:

pip install -U "huggingface_hub[cli]"

Then, you can target the specific file you want:

huggingface-cli download bartowski/Q2.5-MS-Mistoria-72b-v2-GGUF --include "Q2.5-MS-Mistoria-72b-v2-Q4_K_M.gguf" --local-dir ./

If the model is bigger than 50GB, it will have been split into multiple files. In order to download them all to a local folder, run:

huggingface-cli download bartowski/Q2.5-MS-Mistoria-72b-v2-GGUF --include "Q2.5-MS-Mistoria-72b-v2-Q8_0/*" --local-dir ./

You can either specify a new local-dir (Q2.5-MS-Mistoria-72b-v2-Q8_0) or download them all in place (./)

Q4_0_X_X information

Click to view Q4_0_X_X information

These are *NOT* for Metal (Apple) or GPU (nvidia/AMD/intel) offloading, only ARM chips (and certain AVX2/AVX512 CPUs).

If you're using an ARM chip, the Q4_0_X_X quants will have a substantial speedup. Check out Q4_0_4_4 speed comparisons on the original pull request

To check which one would work best for your ARM chip, you can check AArch64 SoC features (thanks EloyOn!).

If you're using a CPU that supports AVX2 or AVX512 (typically server CPUs and AMD's latest Zen5 CPUs) and are not offloading to a GPU, the Q4_0_8_8 may offer a nice speed as well:

Click to view benchmarks on an AVX2 system (EPYC7702)

model	size	params	backend	threads	test	t/s	% (vs Q4_0)
qwen2 3B Q4_0	1.70 GiB	3.09 B	CPU	64	pp512	204.03 ± 1.03	100%
qwen2 3B Q4_0	1.70 GiB	3.09 B	CPU	64	pp1024	282.92 ± 0.19	100%
qwen2 3B Q4_0	1.70 GiB	3.09 B	CPU	64	pp2048	259.49 ± 0.44	100%
qwen2 3B Q4_0	1.70 GiB	3.09 B	CPU	64	tg128	39.12 ± 0.27	100%
qwen2 3B Q4_0	1.70 GiB	3.09 B	CPU	64	tg256	39.31 ± 0.69	100%
qwen2 3B Q4_0	1.70 GiB	3.09 B	CPU	64	tg512	40.52 ± 0.03	100%
qwen2 3B Q4_K_M	1.79 GiB	3.09 B	CPU	64	pp512	301.02 ± 1.74	147%
qwen2 3B Q4_K_M	1.79 GiB	3.09 B	CPU	64	pp1024	287.23 ± 0.20	101%
qwen2 3B Q4_K_M	1.79 GiB	3.09 B	CPU	64	pp2048	262.77 ± 1.81	101%
qwen2 3B Q4_K_M	1.79 GiB	3.09 B	CPU	64	tg128	18.80 ± 0.99	48%
qwen2 3B Q4_K_M	1.79 GiB	3.09 B	CPU	64	tg256	24.46 ± 3.04	83%
qwen2 3B Q4_K_M	1.79 GiB	3.09 B	CPU	64	tg512	36.32 ± 3.59	90%
qwen2 3B Q4_0_8_8	1.69 GiB	3.09 B	CPU	64	pp512	271.71 ± 3.53	133%
qwen2 3B Q4_0_8_8	1.69 GiB	3.09 B	CPU	64	pp1024	279.86 ± 45.63	100%
qwen2 3B Q4_0_8_8	1.69 GiB	3.09 B	CPU	64	pp2048	320.77 ± 5.00	124%
qwen2 3B Q4_0_8_8	1.69 GiB	3.09 B	CPU	64	tg128	43.51 ± 0.05	111%
qwen2 3B Q4_0_8_8	1.69 GiB	3.09 B	CPU	64	tg256	43.35 ± 0.09	110%
qwen2 3B Q4_0_8_8	1.69 GiB	3.09 B	CPU	64	tg512	42.60 ± 0.31	105%

Q4_0_8_8 offers a nice bump to prompt processing and a small bump to text generation

Which file should I choose?

Click here for details

A great write up with charts showing various performances is provided by Artefact2 here

The first thing to figure out is how big a model you can run. To do this, you'll need to figure out how much RAM and/or VRAM you have.

If you want your model running as FAST as possible, you'll want to fit the whole thing on your GPU's VRAM. Aim for a quant with a file size 1-2GB smaller than your GPU's total VRAM.

If you want the absolute maximum quality, add both your system RAM and your GPU's VRAM together, then similarly grab a quant with a file size 1-2GB Smaller than that total.

Next, you'll need to decide if you want to use an 'I-quant' or a 'K-quant'.

If you don't want to think too much, grab one of the K-quants. These are in format 'QX_K_X', like Q5_K_M.

If you want to get more into the weeds, you can check out this extremely useful feature chart:

llama.cpp feature matrix

But basically, if you're aiming for below Q4, and you're running cuBLAS (Nvidia) or rocBLAS (AMD), you should look towards the I-quants. These are in format IQX_X, like IQ3_M. These are newer and offer better performance for their size.

These I-quants can also be used on CPU and Apple Metal, but will be slower than their K-quant equivalent, so speed vs performance is a tradeoff you'll have to decide.

The I-quants are not compatible with Vulcan, which is also AMD, so if you have an AMD card double check if you're using the rocBLAS build or the Vulcan build. At the time of writing this, LM Studio has a preview with ROCm support, and other inference engines have specific builds for ROCm.

Credits

Thank you kalomaze and Dampf for assistance in creating the imatrix calibration dataset.

Thank you ZeroWw for the inspiration to experiment with embed/output.

Want to support my work? Visit my ko-fi page here: https://ko-fi.com/bartowski