From 75850969e32b96955af0714aebbf89461e61819b Mon Sep 17 00:00:00 2001 From: ModelHub XC Date: Mon, 13 Jul 2026 18:21:05 +0800 Subject: [PATCH] =?UTF-8?q?=E5=88=9D=E5=A7=8B=E5=8C=96=E9=A1=B9=E7=9B=AE?= =?UTF-8?q?=EF=BC=8C=E7=94=B1ModelHub=20XC=E7=A4=BE=E5=8C=BA=E6=8F=90?= =?UTF-8?q?=E4=BE=9B=E6=A8=A1=E5=9E=8B?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Model: bartowski/TheDrummer_Anubis-Mini-8B-v1-GGUF Source: Original Platform --- .gitattributes | 73 +++++++++ README.md | 180 ++++++++++++++++++++++ TheDrummer_Anubis-Mini-8B-v1-IQ2_M.gguf | 3 + TheDrummer_Anubis-Mini-8B-v1-IQ3_M.gguf | 3 + TheDrummer_Anubis-Mini-8B-v1-IQ3_XS.gguf | 3 + TheDrummer_Anubis-Mini-8B-v1-IQ3_XXS.gguf | 3 + TheDrummer_Anubis-Mini-8B-v1-IQ4_NL.gguf | 3 + TheDrummer_Anubis-Mini-8B-v1-IQ4_XS.gguf | 3 + TheDrummer_Anubis-Mini-8B-v1-Q2_K.gguf | 3 + TheDrummer_Anubis-Mini-8B-v1-Q2_K_L.gguf | 3 + TheDrummer_Anubis-Mini-8B-v1-Q3_K_L.gguf | 3 + TheDrummer_Anubis-Mini-8B-v1-Q3_K_M.gguf | 3 + TheDrummer_Anubis-Mini-8B-v1-Q3_K_S.gguf | 3 + TheDrummer_Anubis-Mini-8B-v1-Q3_K_XL.gguf | 3 + TheDrummer_Anubis-Mini-8B-v1-Q4_0.gguf | 3 + TheDrummer_Anubis-Mini-8B-v1-Q4_1.gguf | 3 + TheDrummer_Anubis-Mini-8B-v1-Q4_K_L.gguf | 3 + TheDrummer_Anubis-Mini-8B-v1-Q4_K_M.gguf | 3 + TheDrummer_Anubis-Mini-8B-v1-Q4_K_S.gguf | 3 + TheDrummer_Anubis-Mini-8B-v1-Q5_K_L.gguf | 3 + TheDrummer_Anubis-Mini-8B-v1-Q5_K_M.gguf | 3 + TheDrummer_Anubis-Mini-8B-v1-Q5_K_S.gguf | 3 + TheDrummer_Anubis-Mini-8B-v1-Q6_K.gguf | 3 + TheDrummer_Anubis-Mini-8B-v1-Q6_K_L.gguf | 3 + TheDrummer_Anubis-Mini-8B-v1-Q8_0.gguf | 3 + TheDrummer_Anubis-Mini-8B-v1-bf16.gguf | 3 + TheDrummer_Anubis-Mini-8B-v1-imatrix.gguf | 3 + configuration.json | 1 + 28 files changed, 329 insertions(+) create mode 100644 .gitattributes create mode 100644 README.md create mode 100644 TheDrummer_Anubis-Mini-8B-v1-IQ2_M.gguf create mode 100644 TheDrummer_Anubis-Mini-8B-v1-IQ3_M.gguf create mode 100644 TheDrummer_Anubis-Mini-8B-v1-IQ3_XS.gguf create mode 100644 TheDrummer_Anubis-Mini-8B-v1-IQ3_XXS.gguf create mode 100644 TheDrummer_Anubis-Mini-8B-v1-IQ4_NL.gguf create mode 100644 TheDrummer_Anubis-Mini-8B-v1-IQ4_XS.gguf create mode 100644 TheDrummer_Anubis-Mini-8B-v1-Q2_K.gguf create mode 100644 TheDrummer_Anubis-Mini-8B-v1-Q2_K_L.gguf create mode 100644 TheDrummer_Anubis-Mini-8B-v1-Q3_K_L.gguf create mode 100644 TheDrummer_Anubis-Mini-8B-v1-Q3_K_M.gguf create mode 100644 TheDrummer_Anubis-Mini-8B-v1-Q3_K_S.gguf create mode 100644 TheDrummer_Anubis-Mini-8B-v1-Q3_K_XL.gguf create mode 100644 TheDrummer_Anubis-Mini-8B-v1-Q4_0.gguf create mode 100644 TheDrummer_Anubis-Mini-8B-v1-Q4_1.gguf create mode 100644 TheDrummer_Anubis-Mini-8B-v1-Q4_K_L.gguf create mode 100644 TheDrummer_Anubis-Mini-8B-v1-Q4_K_M.gguf create mode 100644 TheDrummer_Anubis-Mini-8B-v1-Q4_K_S.gguf create mode 100644 TheDrummer_Anubis-Mini-8B-v1-Q5_K_L.gguf create mode 100644 TheDrummer_Anubis-Mini-8B-v1-Q5_K_M.gguf create mode 100644 TheDrummer_Anubis-Mini-8B-v1-Q5_K_S.gguf create mode 100644 TheDrummer_Anubis-Mini-8B-v1-Q6_K.gguf create mode 100644 TheDrummer_Anubis-Mini-8B-v1-Q6_K_L.gguf create mode 100644 TheDrummer_Anubis-Mini-8B-v1-Q8_0.gguf create mode 100644 TheDrummer_Anubis-Mini-8B-v1-bf16.gguf create mode 100644 TheDrummer_Anubis-Mini-8B-v1-imatrix.gguf create mode 100644 configuration.json diff --git a/.gitattributes b/.gitattributes new file mode 100644 index 0000000..41cecd8 --- /dev/null +++ b/.gitattributes @@ -0,0 +1,73 @@ +*.7z filter=lfs diff=lfs merge=lfs -text +*.arrow filter=lfs diff=lfs merge=lfs -text +*.bin filter=lfs diff=lfs merge=lfs -text +*.bin.* filter=lfs diff=lfs merge=lfs -text +*.bz2 filter=lfs diff=lfs merge=lfs -text +*.ftz filter=lfs diff=lfs merge=lfs -text +*.gz filter=lfs diff=lfs merge=lfs -text +*.h5 filter=lfs diff=lfs merge=lfs -text +*.joblib filter=lfs diff=lfs merge=lfs -text +*.lfs.* filter=lfs diff=lfs merge=lfs -text +*.model filter=lfs diff=lfs merge=lfs -text +*.msgpack filter=lfs diff=lfs merge=lfs -text +*.onnx filter=lfs diff=lfs merge=lfs -text +*.ot filter=lfs diff=lfs merge=lfs -text +*.parquet filter=lfs diff=lfs merge=lfs -text +*.pb filter=lfs diff=lfs merge=lfs -text +*.pt filter=lfs diff=lfs merge=lfs -text +*.pth filter=lfs diff=lfs merge=lfs -text +*.rar filter=lfs diff=lfs merge=lfs -text +saved_model/**/* filter=lfs diff=lfs merge=lfs -text +*.tar.* filter=lfs diff=lfs merge=lfs -text +*.tflite filter=lfs diff=lfs merge=lfs -text +*.tgz filter=lfs diff=lfs merge=lfs -text +*.xz filter=lfs diff=lfs merge=lfs -text +*.zip filter=lfs diff=lfs merge=lfs -text +*.zstandard filter=lfs diff=lfs merge=lfs -text +*.tfevents* filter=lfs diff=lfs merge=lfs -text +*.db* filter=lfs diff=lfs merge=lfs -text +*.ark* filter=lfs diff=lfs merge=lfs -text +**/*ckpt*data* filter=lfs diff=lfs merge=lfs -text +**/*ckpt*.meta filter=lfs diff=lfs merge=lfs -text +**/*ckpt*.index filter=lfs diff=lfs merge=lfs -text +*.safetensors filter=lfs diff=lfs merge=lfs -text +*.ckpt filter=lfs diff=lfs merge=lfs -text + +*.ggml filter=lfs diff=lfs merge=lfs -text +*.llamafile* filter=lfs diff=lfs merge=lfs -text +*.pt2 filter=lfs diff=lfs merge=lfs -text +*.mlmodel filter=lfs diff=lfs merge=lfs -text +*.npy filter=lfs diff=lfs merge=lfs -text +*.npz filter=lfs diff=lfs merge=lfs -text +*.pickle filter=lfs diff=lfs merge=lfs -text +*.pkl filter=lfs diff=lfs merge=lfs -text +*.tar filter=lfs diff=lfs merge=lfs -text +*.wasm filter=lfs diff=lfs merge=lfs -text +*.zst filter=lfs diff=lfs merge=lfs -text +*tfevents* filter=lfs diff=lfs merge=lfs -text + +TheDrummer_Anubis-Mini-8B-v1-IQ2_M.gguf filter=lfs diff=lfs merge=lfs -text +TheDrummer_Anubis-Mini-8B-v1-Q3_K_M.gguf filter=lfs diff=lfs merge=lfs -text +TheDrummer_Anubis-Mini-8B-v1-IQ4_XS.gguf filter=lfs diff=lfs merge=lfs -text +TheDrummer_Anubis-Mini-8B-v1-Q3_K_S.gguf filter=lfs diff=lfs merge=lfs -text +TheDrummer_Anubis-Mini-8B-v1-Q5_K_L.gguf filter=lfs diff=lfs merge=lfs -text +TheDrummer_Anubis-Mini-8B-v1-IQ3_M.gguf filter=lfs diff=lfs merge=lfs -text +TheDrummer_Anubis-Mini-8B-v1-bf16.gguf filter=lfs diff=lfs merge=lfs -text +TheDrummer_Anubis-Mini-8B-v1-Q5_K_M.gguf filter=lfs diff=lfs merge=lfs -text +TheDrummer_Anubis-Mini-8B-v1-imatrix.gguf filter=lfs diff=lfs merge=lfs -text +TheDrummer_Anubis-Mini-8B-v1-Q2_K.gguf filter=lfs diff=lfs merge=lfs -text +TheDrummer_Anubis-Mini-8B-v1-Q4_K_L.gguf filter=lfs diff=lfs merge=lfs -text +TheDrummer_Anubis-Mini-8B-v1-Q4_K_S.gguf filter=lfs diff=lfs merge=lfs -text +TheDrummer_Anubis-Mini-8B-v1-IQ3_XS.gguf filter=lfs diff=lfs merge=lfs -text +TheDrummer_Anubis-Mini-8B-v1-Q2_K_L.gguf filter=lfs diff=lfs merge=lfs -text +TheDrummer_Anubis-Mini-8B-v1-Q3_K_L.gguf filter=lfs diff=lfs merge=lfs -text +TheDrummer_Anubis-Mini-8B-v1-Q5_K_S.gguf filter=lfs diff=lfs merge=lfs -text +TheDrummer_Anubis-Mini-8B-v1-Q4_0.gguf filter=lfs diff=lfs merge=lfs -text +TheDrummer_Anubis-Mini-8B-v1-Q3_K_XL.gguf filter=lfs diff=lfs merge=lfs -text +TheDrummer_Anubis-Mini-8B-v1-Q6_K.gguf filter=lfs diff=lfs merge=lfs -text +TheDrummer_Anubis-Mini-8B-v1-Q6_K_L.gguf filter=lfs diff=lfs merge=lfs -text +TheDrummer_Anubis-Mini-8B-v1-Q4_K_M.gguf filter=lfs diff=lfs merge=lfs -text +TheDrummer_Anubis-Mini-8B-v1-Q8_0.gguf filter=lfs diff=lfs merge=lfs -text +TheDrummer_Anubis-Mini-8B-v1-IQ4_NL.gguf filter=lfs diff=lfs merge=lfs -text +TheDrummer_Anubis-Mini-8B-v1-IQ3_XXS.gguf filter=lfs diff=lfs merge=lfs -text +TheDrummer_Anubis-Mini-8B-v1-Q4_1.gguf filter=lfs diff=lfs merge=lfs -text \ No newline at end of file diff --git a/README.md b/README.md new file mode 100644 index 0000000..bfb25c7 --- /dev/null +++ b/README.md @@ -0,0 +1,180 @@ +--- +quantized_by: bartowski +pipeline_tag: text-generation +base_model: TheDrummer/Anubis-Mini-8B-v1 +base_model_relation: quantized +--- + +## Llamacpp imatrix Quantizations of Anubis-Mini-8B-v1 by TheDrummer + +Using llama.cpp release b8388 for quantization. + +Original model: https://huggingface.co/TheDrummer/Anubis-Mini-8B-v1 + +All quants made using imatrix option with dataset from [here](https://gist.github.com/bartowski1182/82ae9b520227f57d79ba04add13d0d0d) + +Run them in your choice of tools: + +- [llama.cpp](https://github.com/ggml-org/llama.cpp) +- [LM Studio](https://lmstudio.ai/) +- [koboldcpp](https://github.com/LostRuins/koboldcpp) +- [Jan AI](https://www.jan.ai/) +- [Text Generation Web UI](https://github.com/oobabooga/text-generation-webui) +- [LoLLMs](https://github.com/ParisNeo/lollms) + +Note: if it's a newly supported model, you may need to wait for an update from the developers. + +## Prompt format + +``` +<|begin_of_text|><|start_header_id|>system<|end_header_id|> + +Cutting Knowledge Date: December 2023 +Today Date: 30 Dec 2025 + +{system_prompt}<|eot_id|><|start_header_id|>user<|end_header_id|> + +{prompt}<|eot_id|><|start_header_id|>assistant<|end_header_id|> +``` + +## Download a file (not the whole branch) from below: + +| Filename | Quant type | File Size | Split | Description | +| -------- | ---------- | --------- | ----- | ----------- | +| [Anubis-Mini-8B-v1-bf16.gguf](https://huggingface.co/bartowski/TheDrummer_Anubis-Mini-8B-v1-GGUF/blob/main/TheDrummer_Anubis-Mini-8B-v1-bf16.gguf) | bf16 | 16.07GB | false | Full BF16 weights. | +| [Anubis-Mini-8B-v1-Q8_0.gguf](https://huggingface.co/bartowski/TheDrummer_Anubis-Mini-8B-v1-GGUF/blob/main/TheDrummer_Anubis-Mini-8B-v1-Q8_0.gguf) | Q8_0 | 8.54GB | false | Extremely high quality, generally unneeded but max available quant. | +| [Anubis-Mini-8B-v1-Q6_K_L.gguf](https://huggingface.co/bartowski/TheDrummer_Anubis-Mini-8B-v1-GGUF/blob/main/TheDrummer_Anubis-Mini-8B-v1-Q6_K_L.gguf) | Q6_K_L | 7.09GB | false | Uses Q8_0 for embed and output weights. Very high quality, near perfect, *recommended*. | +| [Anubis-Mini-8B-v1-Q6_K.gguf](https://huggingface.co/bartowski/TheDrummer_Anubis-Mini-8B-v1-GGUF/blob/main/TheDrummer_Anubis-Mini-8B-v1-Q6_K.gguf) | Q6_K | 6.84GB | false | Very high quality, near perfect, *recommended*. | +| [Anubis-Mini-8B-v1-Q5_K_L.gguf](https://huggingface.co/bartowski/TheDrummer_Anubis-Mini-8B-v1-GGUF/blob/main/TheDrummer_Anubis-Mini-8B-v1-Q5_K_L.gguf) | Q5_K_L | 6.18GB | false | Uses Q8_0 for embed and output weights. High quality, *recommended*. | +| [Anubis-Mini-8B-v1-Q5_K_M.gguf](https://huggingface.co/bartowski/TheDrummer_Anubis-Mini-8B-v1-GGUF/blob/main/TheDrummer_Anubis-Mini-8B-v1-Q5_K_M.gguf) | Q5_K_M | 5.86GB | false | High quality, *recommended*. | +| [Anubis-Mini-8B-v1-Q5_K_S.gguf](https://huggingface.co/bartowski/TheDrummer_Anubis-Mini-8B-v1-GGUF/blob/main/TheDrummer_Anubis-Mini-8B-v1-Q5_K_S.gguf) | Q5_K_S | 5.63GB | false | High quality, *recommended*. | +| [Anubis-Mini-8B-v1-Q4_K_L.gguf](https://huggingface.co/bartowski/TheDrummer_Anubis-Mini-8B-v1-GGUF/blob/main/TheDrummer_Anubis-Mini-8B-v1-Q4_K_L.gguf) | Q4_K_L | 5.46GB | false | Uses Q8_0 for embed and output weights. Good quality, *recommended*. | +| [Anubis-Mini-8B-v1-Q4_1.gguf](https://huggingface.co/bartowski/TheDrummer_Anubis-Mini-8B-v1-GGUF/blob/main/TheDrummer_Anubis-Mini-8B-v1-Q4_1.gguf) | Q4_1 | 5.18GB | false | Legacy format, similar performance to Q4_K_S but with improved tokens/watt on Apple silicon. | +| [Anubis-Mini-8B-v1-Q4_K_M.gguf](https://huggingface.co/bartowski/TheDrummer_Anubis-Mini-8B-v1-GGUF/blob/main/TheDrummer_Anubis-Mini-8B-v1-Q4_K_M.gguf) | Q4_K_M | 5.07GB | false | Good quality, default size for most use cases, *recommended*. | +| [Anubis-Mini-8B-v1-Q3_K_XL.gguf](https://huggingface.co/bartowski/TheDrummer_Anubis-Mini-8B-v1-GGUF/blob/main/TheDrummer_Anubis-Mini-8B-v1-Q3_K_XL.gguf) | Q3_K_XL | 4.82GB | false | Uses Q8_0 for embed and output weights. Lower quality but usable, good for low RAM availability. | +| [Anubis-Mini-8B-v1-Q4_K_S.gguf](https://huggingface.co/bartowski/TheDrummer_Anubis-Mini-8B-v1-GGUF/blob/main/TheDrummer_Anubis-Mini-8B-v1-Q4_K_S.gguf) | Q4_K_S | 4.77GB | false | Slightly lower quality with more space savings, *recommended*. | +| [Anubis-Mini-8B-v1-IQ4_NL.gguf](https://huggingface.co/bartowski/TheDrummer_Anubis-Mini-8B-v1-GGUF/blob/main/TheDrummer_Anubis-Mini-8B-v1-IQ4_NL.gguf) | IQ4_NL | 4.77GB | false | Similar to IQ4_XS, but slightly larger. Offers online repacking for ARM CPU inference. | +| [Anubis-Mini-8B-v1-Q4_0.gguf](https://huggingface.co/bartowski/TheDrummer_Anubis-Mini-8B-v1-GGUF/blob/main/TheDrummer_Anubis-Mini-8B-v1-Q4_0.gguf) | Q4_0 | 4.75GB | false | Legacy format, offers online repacking for ARM and AVX CPU inference. | +| [Anubis-Mini-8B-v1-IQ4_XS.gguf](https://huggingface.co/bartowski/TheDrummer_Anubis-Mini-8B-v1-GGUF/blob/main/TheDrummer_Anubis-Mini-8B-v1-IQ4_XS.gguf) | IQ4_XS | 4.54GB | false | Decent quality, smaller than Q4_K_S with similar performance, *recommended*. | +| [Anubis-Mini-8B-v1-Q3_K_L.gguf](https://huggingface.co/bartowski/TheDrummer_Anubis-Mini-8B-v1-GGUF/blob/main/TheDrummer_Anubis-Mini-8B-v1-Q3_K_L.gguf) | Q3_K_L | 4.36GB | false | Lower quality but usable, good for low RAM availability. | +| [Anubis-Mini-8B-v1-Q3_K_M.gguf](https://huggingface.co/bartowski/TheDrummer_Anubis-Mini-8B-v1-GGUF/blob/main/TheDrummer_Anubis-Mini-8B-v1-Q3_K_M.gguf) | Q3_K_M | 4.13GB | false | Low quality. | +| [Anubis-Mini-8B-v1-IQ3_M.gguf](https://huggingface.co/bartowski/TheDrummer_Anubis-Mini-8B-v1-GGUF/blob/main/TheDrummer_Anubis-Mini-8B-v1-IQ3_M.gguf) | IQ3_M | 3.91GB | false | Medium-low quality, new method with decent performance comparable to Q3_K_M. | +| [Anubis-Mini-8B-v1-Q2_K_L.gguf](https://huggingface.co/bartowski/TheDrummer_Anubis-Mini-8B-v1-GGUF/blob/main/TheDrummer_Anubis-Mini-8B-v1-Q2_K_L.gguf) | Q2_K_L | 3.82GB | false | Uses Q8_0 for embed and output weights. Very low quality but surprisingly usable. | +| [Anubis-Mini-8B-v1-Q3_K_S.gguf](https://huggingface.co/bartowski/TheDrummer_Anubis-Mini-8B-v1-GGUF/blob/main/TheDrummer_Anubis-Mini-8B-v1-Q3_K_S.gguf) | Q3_K_S | 3.77GB | false | Low quality, not recommended. | +| [Anubis-Mini-8B-v1-IQ3_XS.gguf](https://huggingface.co/bartowski/TheDrummer_Anubis-Mini-8B-v1-GGUF/blob/main/TheDrummer_Anubis-Mini-8B-v1-IQ3_XS.gguf) | IQ3_XS | 3.66GB | false | Lower quality, new method with decent performance, slightly better than Q3_K_S. | +| [Anubis-Mini-8B-v1-IQ3_XXS.gguf](https://huggingface.co/bartowski/TheDrummer_Anubis-Mini-8B-v1-GGUF/blob/main/TheDrummer_Anubis-Mini-8B-v1-IQ3_XXS.gguf) | IQ3_XXS | 3.43GB | false | Lower quality, new method with decent performance, comparable to Q3 quants. | +| [Anubis-Mini-8B-v1-Q2_K.gguf](https://huggingface.co/bartowski/TheDrummer_Anubis-Mini-8B-v1-GGUF/blob/main/TheDrummer_Anubis-Mini-8B-v1-Q2_K.gguf) | Q2_K | 3.31GB | false | Very low quality but surprisingly usable. | +| [Anubis-Mini-8B-v1-IQ2_M.gguf](https://huggingface.co/bartowski/TheDrummer_Anubis-Mini-8B-v1-GGUF/blob/main/TheDrummer_Anubis-Mini-8B-v1-IQ2_M.gguf) | IQ2_M | 3.26GB | false | Relatively low quality, uses SOTA techniques to be surprisingly 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/TheDrummer_Anubis-Mini-8B-v1-GGUF --include "TheDrummer_Anubis-Mini-8B-v1-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/TheDrummer_Anubis-Mini-8B-v1-GGUF --include "TheDrummer_Anubis-Mini-8B-v1-Q8_0/*" --local-dir ./ +``` + +You can either specify a new local-dir (TheDrummer_Anubis-Mini-8B-v1-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](https://github.com/ggml-org/llama.cpp/pull/9921). 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](https://github.com/ggml-org/llama.cpp/releases/tag/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](https://github.com/ggml-org/llama.cpp/pull/10541) 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](https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9) + +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](https://github.com/ggml-org/llama.cpp/wiki/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, but will be slower than their K-quant equivalent, so speed vs performance is a tradeoff you'll have to decide. + +
+ +## 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. + +Thank you to LM Studio for sponsoring my work. + +Want to support my work? Visit my ko-fi page here: https://ko-fi.com/bartowski diff --git a/TheDrummer_Anubis-Mini-8B-v1-IQ2_M.gguf b/TheDrummer_Anubis-Mini-8B-v1-IQ2_M.gguf new file mode 100644 index 0000000..203ca68 --- /dev/null +++ b/TheDrummer_Anubis-Mini-8B-v1-IQ2_M.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:362467471fb671268583b81175321931246e1254df6d8dfa89965c688a8fcf52 +size 3256572096 diff --git a/TheDrummer_Anubis-Mini-8B-v1-IQ3_M.gguf b/TheDrummer_Anubis-Mini-8B-v1-IQ3_M.gguf new file mode 100644 index 0000000..d268bf3 --- /dev/null +++ b/TheDrummer_Anubis-Mini-8B-v1-IQ3_M.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:849eec79345d270649922480a72567e587b0db857df5c15b49a8facbf9fb850a +size 3906468576 diff --git a/TheDrummer_Anubis-Mini-8B-v1-IQ3_XS.gguf b/TheDrummer_Anubis-Mini-8B-v1-IQ3_XS.gguf new file mode 100644 index 0000000..37d8000 --- /dev/null +++ b/TheDrummer_Anubis-Mini-8B-v1-IQ3_XS.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:813210956a4fb9dfe6803887521bdf2d2c1d50aeb14c3851ae5fd682a4f5a024 +size 3655072480 diff --git a/TheDrummer_Anubis-Mini-8B-v1-IQ3_XXS.gguf b/TheDrummer_Anubis-Mini-8B-v1-IQ3_XXS.gguf new file mode 100644 index 0000000..f3a3086 --- /dev/null +++ b/TheDrummer_Anubis-Mini-8B-v1-IQ3_XXS.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:184c9879728945006dacfffb322176a706cc5feb0ed295b6e10f0efd8bd26ede +size 3434305728 diff --git a/TheDrummer_Anubis-Mini-8B-v1-IQ4_NL.gguf b/TheDrummer_Anubis-Mini-8B-v1-IQ4_NL.gguf new file mode 100644 index 0000000..66288ed --- /dev/null +++ b/TheDrummer_Anubis-Mini-8B-v1-IQ4_NL.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7f211ebe94cd2e935b874888afc81d1b45d4059d70b62ab1bec1c1f492e25ba6 +size 4765031680 diff --git a/TheDrummer_Anubis-Mini-8B-v1-IQ4_XS.gguf b/TheDrummer_Anubis-Mini-8B-v1-IQ4_XS.gguf new file mode 100644 index 0000000..216003b --- /dev/null +++ b/TheDrummer_Anubis-Mini-8B-v1-IQ4_XS.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b11124606747c3adc46e1e8cf391b9c7d01ed9176f5bd3544f88b58519846e4f +size 4543093888 diff --git a/TheDrummer_Anubis-Mini-8B-v1-Q2_K.gguf b/TheDrummer_Anubis-Mini-8B-v1-Q2_K.gguf new file mode 100644 index 0000000..b98a374 --- /dev/null +++ b/TheDrummer_Anubis-Mini-8B-v1-Q2_K.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ad7c3b8234cfe87872bf849a90179c396479d80be16ab37893935f246fffdb07 +size 3311262016 diff --git a/TheDrummer_Anubis-Mini-8B-v1-Q2_K_L.gguf b/TheDrummer_Anubis-Mini-8B-v1-Q2_K_L.gguf new file mode 100644 index 0000000..424cb70 --- /dev/null +++ b/TheDrummer_Anubis-Mini-8B-v1-Q2_K_L.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:475bb88b3cb933a06fc1dfb06c2bf0bb2af2848016f65572a8633a6c7e6e75d7 +size 3824290016 diff --git a/TheDrummer_Anubis-Mini-8B-v1-Q3_K_L.gguf b/TheDrummer_Anubis-Mini-8B-v1-Q3_K_L.gguf new file mode 100644 index 0000000..17f0563 --- /dev/null +++ b/TheDrummer_Anubis-Mini-8B-v1-Q3_K_L.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b21ad246f76eac6b2d90c6f45fe9be8fd4c46fca262feb697da6bdf5348e65db +size 4359715552 diff --git a/TheDrummer_Anubis-Mini-8B-v1-Q3_K_M.gguf b/TheDrummer_Anubis-Mini-8B-v1-Q3_K_M.gguf new file mode 100644 index 0000000..a96b993 --- /dev/null +++ b/TheDrummer_Anubis-Mini-8B-v1-Q3_K_M.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3e19491a0ae301b24e3042b458ef7deb3414cf08956f60dc1a0598f60ca501c6 +size 4131125984 diff --git a/TheDrummer_Anubis-Mini-8B-v1-Q3_K_S.gguf b/TheDrummer_Anubis-Mini-8B-v1-Q3_K_S.gguf new file mode 100644 index 0000000..3169556 --- /dev/null +++ b/TheDrummer_Anubis-Mini-8B-v1-Q3_K_S.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ec44d1fba8f29fb9741783715f3c3867ba55bbda6a8f224c4cc8ab94885eff4d +size 3769367264 diff --git a/TheDrummer_Anubis-Mini-8B-v1-Q3_K_XL.gguf b/TheDrummer_Anubis-Mini-8B-v1-Q3_K_XL.gguf new file mode 100644 index 0000000..2a0e6f3 --- /dev/null +++ b/TheDrummer_Anubis-Mini-8B-v1-Q3_K_XL.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:771d938cb8c8b5f7b6e56392fa0ce69a2b0a1e7b26cbf71692d5d508a6214896 +size 4819388640 diff --git a/TheDrummer_Anubis-Mini-8B-v1-Q4_0.gguf b/TheDrummer_Anubis-Mini-8B-v1-Q4_0.gguf new file mode 100644 index 0000000..52cc1a7 --- /dev/null +++ b/TheDrummer_Anubis-Mini-8B-v1-Q4_0.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6d1a58d810e70f378f535dea984bd9077400ad17271ad8d51ba93f26d9475cea +size 4745108736 diff --git a/TheDrummer_Anubis-Mini-8B-v1-Q4_1.gguf b/TheDrummer_Anubis-Mini-8B-v1-Q4_1.gguf new file mode 100644 index 0000000..fd1654f --- /dev/null +++ b/TheDrummer_Anubis-Mini-8B-v1-Q4_1.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e67172ec52842b84bc9bc35c71c43c713656c2d52f134d00da331e66b0f5871b +size 5182692864 diff --git a/TheDrummer_Anubis-Mini-8B-v1-Q4_K_L.gguf b/TheDrummer_Anubis-Mini-8B-v1-Q4_K_L.gguf new file mode 100644 index 0000000..34488ed --- /dev/null +++ b/TheDrummer_Anubis-Mini-8B-v1-Q4_K_L.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b6f55695c1bdb88b438164471be21c2d4324a4fabe34299bf490316fc37505fc +size 5463738592 diff --git a/TheDrummer_Anubis-Mini-8B-v1-Q4_K_M.gguf b/TheDrummer_Anubis-Mini-8B-v1-Q4_K_M.gguf new file mode 100644 index 0000000..1825846 --- /dev/null +++ b/TheDrummer_Anubis-Mini-8B-v1-Q4_K_M.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:411c017e9c39b97ecf8373e34d348bb4bcbc4ef3aa761cb109bf5af0d1ef444c +size 5073837312 diff --git a/TheDrummer_Anubis-Mini-8B-v1-Q4_K_S.gguf b/TheDrummer_Anubis-Mini-8B-v1-Q4_K_S.gguf new file mode 100644 index 0000000..3a8fbcd --- /dev/null +++ b/TheDrummer_Anubis-Mini-8B-v1-Q4_K_S.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d2370a5dfe527d92713695c6562c19ae7a9aefeaca1325e2b16509095319f702 +size 4768177408 diff --git a/TheDrummer_Anubis-Mini-8B-v1-Q5_K_L.gguf b/TheDrummer_Anubis-Mini-8B-v1-Q5_K_L.gguf new file mode 100644 index 0000000..e8917ed --- /dev/null +++ b/TheDrummer_Anubis-Mini-8B-v1-Q5_K_L.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2bd9a5b4da893d15688369f2f98ce2f68b5a0d0a7dfc761e45b48a47c5fc5310 +size 6182013152 diff --git a/TheDrummer_Anubis-Mini-8B-v1-Q5_K_M.gguf b/TheDrummer_Anubis-Mini-8B-v1-Q5_K_M.gguf new file mode 100644 index 0000000..6192c76 --- /dev/null +++ b/TheDrummer_Anubis-Mini-8B-v1-Q5_K_M.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ce49b8d4bd7cb19d02a86e68b1dc202596f696f898b882acc65b6cd2c018277d +size 5857779456 diff --git a/TheDrummer_Anubis-Mini-8B-v1-Q5_K_S.gguf b/TheDrummer_Anubis-Mini-8B-v1-Q5_K_S.gguf new file mode 100644 index 0000000..9308b8e --- /dev/null +++ b/TheDrummer_Anubis-Mini-8B-v1-Q5_K_S.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f93f8e23c0fd960b6ba3fd8f6dd573f3562cfed19c3b97602c115e059fd72fb9 +size 5634957056 diff --git a/TheDrummer_Anubis-Mini-8B-v1-Q6_K.gguf b/TheDrummer_Anubis-Mini-8B-v1-Q6_K.gguf new file mode 100644 index 0000000..0c3d6e3 --- /dev/null +++ b/TheDrummer_Anubis-Mini-8B-v1-Q6_K.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c34cb097d672f6855ec818295ee5794b215accae5f4190848dd5fe83c4d7edfe +size 6839812384 diff --git a/TheDrummer_Anubis-Mini-8B-v1-Q6_K_L.gguf b/TheDrummer_Anubis-Mini-8B-v1-Q6_K_L.gguf new file mode 100644 index 0000000..4bcd929 --- /dev/null +++ b/TheDrummer_Anubis-Mini-8B-v1-Q6_K_L.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:eed1cec0efc255dd77e125e0e563ebec8c61356cffc33fb4530082f6d6a9ac0e +size 7094274272 diff --git a/TheDrummer_Anubis-Mini-8B-v1-Q8_0.gguf b/TheDrummer_Anubis-Mini-8B-v1-Q8_0.gguf new file mode 100644 index 0000000..8b9cee2 --- /dev/null +++ b/TheDrummer_Anubis-Mini-8B-v1-Q8_0.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1cc61a350b61b9b17c0cb486cb2554b575eb0376e7d5bc8ad98a6736c0df3ef5 +size 8540784864 diff --git a/TheDrummer_Anubis-Mini-8B-v1-bf16.gguf b/TheDrummer_Anubis-Mini-8B-v1-bf16.gguf new file mode 100644 index 0000000..e4ea0ce --- /dev/null +++ b/TheDrummer_Anubis-Mini-8B-v1-bf16.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e0c31458c11fe6bf632a285f08fc72a7a8cb51ce11a597d1d0d33e3dfe6ad1c6 +size 16068912576 diff --git a/TheDrummer_Anubis-Mini-8B-v1-imatrix.gguf b/TheDrummer_Anubis-Mini-8B-v1-imatrix.gguf new file mode 100644 index 0000000..2a29137 --- /dev/null +++ b/TheDrummer_Anubis-Mini-8B-v1-imatrix.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6d4234477480c67f460327ba50b019a555eab0d696949d195c0683f45fa424b4 +size 5015232 diff --git a/configuration.json b/configuration.json new file mode 100644 index 0000000..bbeeda1 --- /dev/null +++ b/configuration.json @@ -0,0 +1 @@ +{"framework": "pytorch", "task": "text-generation", "allow_remote": true} \ No newline at end of file