ozone-ai_0x-lite-GGUFオープンソースモデル - 中国語と英語のテキスト生成をサポート、多様な量子化で異なるハードウェアに適合

ホーム

Ozone Ai 0x Lite GGUF

bartowskiによって開発

ozone-ai/0x-liteモデルに基づく量子化バージョンで、中英テキスト生成タスクをサポートし、llama.cppを使用したimatrix量子化を採用し、さまざまなハードウェア要件に対応する複数の量子化オプションを提供します。

大規模言語モデル複数言語対応オープンソースライセンス:Apache-2.0 #多言語対話 #効率的な量子化 #軽量展開

ダウンロード数 220

リリース時間 : 2/16/2025

モデル概要

これは量子化されたテキスト生成モデルで、中英をサポートし、さまざまな自然言語処理タスクに適しています。モデルは最適化されており、異なるハードウェア構成で効率的に動作します。

モデル特徴

多様な量子化オプション

F16からIQ2_Sまでのさまざまな量子化バージョンを提供し、異なるハードウェア構成と性能要件を満たします。

imatrix量子化技術

llama.cppのimatrixオプションを使用して量子化し、モデル性能を最適化します。

クロスプラットフォームサポート

LM Studio、llama.cpp、およびllama.cppベースのプロジェクトで実行可能です。

中英サポート

中英テキスト生成タスクのサポートを特に最適化しています。

モデル能力

テキスト生成

自然言語理解

マルチターン対話

使用事例

対話システム

インテリジェントカスタマーサポート

顧客の質問に自動返信する対話システムの構築に使用されます。

コンテンツ作成

記事生成

さまざまなタイプの中英記事の作成を支援します。

quantized_by: bartowski pipeline_tag: text-generation license: apache-2.0 base_model: ozone-ai/0x-lite language:

en
zh datasets:
lmsys/lmsys-chat-1m

Llamacpp imatrix Quantizations of 0x-lite by ozone-ai

Using llama.cpp release b4688 for quantization.

Original model: https://huggingface.co/ozone-ai/0x-lite

All quants made using imatrix option with dataset from here

Run them in LM Studio

Run them directly with llama.cpp, or any other llama.cpp based project

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
0x-lite-f16.gguf	f16	29.55GB	false	Full F16 weights.
0x-lite-Q8_0.gguf	Q8_0	15.70GB	false	Extremely high quality, generally unneeded but max available quant.
0x-lite-Q6_K_L.gguf	Q6_K_L	12.50GB	false	Uses Q8_0 for embed and output weights. Very high quality, near perfect, recommended.
0x-lite-Q6_K.gguf	Q6_K	12.12GB	false	Very high quality, near perfect, recommended.
0x-lite-Q5_K_L.gguf	Q5_K_L	10.99GB	false	Uses Q8_0 for embed and output weights. High quality, recommended.
0x-lite-Q5_K_M.gguf	Q5_K_M	10.51GB	false	High quality, recommended.
0x-lite-Q5_K_S.gguf	Q5_K_S	10.27GB	false	High quality, recommended.
0x-lite-Q4_K_L.gguf	Q4_K_L	9.57GB	false	Uses Q8_0 for embed and output weights. Good quality, recommended.
0x-lite-Q4_1.gguf	Q4_1	9.39GB	false	Legacy format, similar performance to Q4_K_S but with improved tokens/watt on Apple silicon.
0x-lite-Q4_K_M.gguf	Q4_K_M	8.99GB	false	Good quality, default size for most use cases, recommended.
0x-lite-Q3_K_XL.gguf	Q3_K_XL	8.61GB	false	Uses Q8_0 for embed and output weights. Lower quality but usable, good for low RAM availability.
0x-lite-Q4_K_S.gguf	Q4_K_S	8.57GB	false	Slightly lower quality with more space savings, recommended.
0x-lite-IQ4_NL.gguf	IQ4_NL	8.55GB	false	Similar to IQ4_XS, but slightly larger. Offers online repacking for ARM CPU inference.
0x-lite-Q4_0.gguf	Q4_0	8.54GB	false	Legacy format, offers online repacking for ARM and AVX CPU inference.
0x-lite-IQ4_XS.gguf	IQ4_XS	8.12GB	false	Decent quality, smaller than Q4_K_S with similar performance, recommended.
0x-lite-Q3_K_L.gguf	Q3_K_L	7.92GB	false	Lower quality but usable, good for low RAM availability.
0x-lite-Q3_K_M.gguf	Q3_K_M	7.34GB	false	Low quality.
0x-lite-IQ3_M.gguf	IQ3_M	6.92GB	false	Medium-low quality, new method with decent performance comparable to Q3_K_M.
0x-lite-Q3_K_S.gguf	Q3_K_S	6.66GB	false	Low quality, not recommended.
0x-lite-Q2_K_L.gguf	Q2_K_L	6.53GB	false	Uses Q8_0 for embed and output weights. Very low quality but surprisingly usable.
0x-lite-IQ3_XS.gguf	IQ3_XS	6.38GB	false	Lower quality, new method with decent performance, slightly better than Q3_K_S.
0x-lite-IQ3_XXS.gguf	IQ3_XXS	5.95GB	false	Lower quality, new method with decent performance, comparable to Q3 quants.
0x-lite-Q2_K.gguf	Q2_K	5.77GB	false	Very low quality but surprisingly usable.
0x-lite-IQ2_M.gguf	IQ2_M	5.36GB	false	Relatively low quality, uses SOTA techniques to be surprisingly usable.
0x-lite-IQ2_S.gguf	IQ2_S	5.00GB	false	Low quality, uses SOTA techniques to be usable.

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/ozone-ai_0x-lite-GGUF --include "ozone-ai_0x-lite-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/ozone-ai_0x-lite-GGUF --include "ozone-ai_0x-lite-Q8_0/*" --local-dir ./

You can either specify a new local-dir (ozone-ai_0x-lite-Q8_0) or download them all in place (./)

ARM/AVX information

Previously, you would download Q4_0_4_4/4_8/8_8, and these would have their weights interleaved in memory in order to improve performance on ARM and AVX machines by loading up more data in one pass.

Now, however, there is something called "online repacking" for weights. details in this PR. If you use Q4_0 and your hardware would benefit from repacking weights, it will do it automatically on the fly.

As of llama.cpp build b4282 you will not be able to run the Q4_0_X_X files and will instead need to use Q4_0.

Additionally, if you want to get slightly better quality for , you can use IQ4_NL thanks to this PR which will also repack the weights for ARM, though only the 4_4 for now. The loading time may be slower but it will result in an overall speed incrase.

Click to view Q4_0_X_X information (deprecated

I'm keeping this section to show the potential theoretical uplift in performance from using the Q4_0 with online repacking.

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.