From 27cb49f9e20d377aecdbac2b9346386298c0e376 Mon Sep 17 00:00:00 2001 From: ModelHub XC Date: Wed, 15 Jul 2026 05:23:10 +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_Rocinante-X-12B-v1-GGUF Source: Original Platform --- .gitattributes | 61 +++++++ README.md | 176 +++++++++++++++++++++ TheDrummer_Rocinante-X-12B-v1-IQ2_M.gguf | 3 + TheDrummer_Rocinante-X-12B-v1-IQ2_S.gguf | 3 + TheDrummer_Rocinante-X-12B-v1-IQ3_M.gguf | 3 + TheDrummer_Rocinante-X-12B-v1-IQ3_XS.gguf | 3 + TheDrummer_Rocinante-X-12B-v1-IQ3_XXS.gguf | 3 + TheDrummer_Rocinante-X-12B-v1-IQ4_NL.gguf | 3 + TheDrummer_Rocinante-X-12B-v1-IQ4_XS.gguf | 3 + TheDrummer_Rocinante-X-12B-v1-Q2_K.gguf | 3 + TheDrummer_Rocinante-X-12B-v1-Q2_K_L.gguf | 3 + TheDrummer_Rocinante-X-12B-v1-Q3_K_L.gguf | 3 + TheDrummer_Rocinante-X-12B-v1-Q3_K_M.gguf | 3 + TheDrummer_Rocinante-X-12B-v1-Q3_K_S.gguf | 3 + TheDrummer_Rocinante-X-12B-v1-Q3_K_XL.gguf | 3 + TheDrummer_Rocinante-X-12B-v1-Q4_0.gguf | 3 + TheDrummer_Rocinante-X-12B-v1-Q4_1.gguf | 3 + TheDrummer_Rocinante-X-12B-v1-Q4_K_L.gguf | 3 + TheDrummer_Rocinante-X-12B-v1-Q4_K_M.gguf | 3 + TheDrummer_Rocinante-X-12B-v1-Q4_K_S.gguf | 3 + TheDrummer_Rocinante-X-12B-v1-Q5_K_L.gguf | 3 + TheDrummer_Rocinante-X-12B-v1-Q5_K_M.gguf | 3 + TheDrummer_Rocinante-X-12B-v1-Q5_K_S.gguf | 3 + TheDrummer_Rocinante-X-12B-v1-Q6_K.gguf | 3 + TheDrummer_Rocinante-X-12B-v1-Q6_K_L.gguf | 3 + TheDrummer_Rocinante-X-12B-v1-Q8_0.gguf | 3 + TheDrummer_Rocinante-X-12B-v1-bf16.gguf | 3 + TheDrummer_Rocinante-X-12B-v1-imatrix.gguf | 3 + 28 files changed, 315 insertions(+) create mode 100644 .gitattributes create mode 100644 README.md create mode 100644 TheDrummer_Rocinante-X-12B-v1-IQ2_M.gguf create mode 100644 TheDrummer_Rocinante-X-12B-v1-IQ2_S.gguf create mode 100644 TheDrummer_Rocinante-X-12B-v1-IQ3_M.gguf create mode 100644 TheDrummer_Rocinante-X-12B-v1-IQ3_XS.gguf create mode 100644 TheDrummer_Rocinante-X-12B-v1-IQ3_XXS.gguf create mode 100644 TheDrummer_Rocinante-X-12B-v1-IQ4_NL.gguf create mode 100644 TheDrummer_Rocinante-X-12B-v1-IQ4_XS.gguf create mode 100644 TheDrummer_Rocinante-X-12B-v1-Q2_K.gguf create mode 100644 TheDrummer_Rocinante-X-12B-v1-Q2_K_L.gguf create mode 100644 TheDrummer_Rocinante-X-12B-v1-Q3_K_L.gguf create mode 100644 TheDrummer_Rocinante-X-12B-v1-Q3_K_M.gguf create mode 100644 TheDrummer_Rocinante-X-12B-v1-Q3_K_S.gguf create mode 100644 TheDrummer_Rocinante-X-12B-v1-Q3_K_XL.gguf create mode 100644 TheDrummer_Rocinante-X-12B-v1-Q4_0.gguf create mode 100644 TheDrummer_Rocinante-X-12B-v1-Q4_1.gguf create mode 100644 TheDrummer_Rocinante-X-12B-v1-Q4_K_L.gguf create mode 100644 TheDrummer_Rocinante-X-12B-v1-Q4_K_M.gguf create mode 100644 TheDrummer_Rocinante-X-12B-v1-Q4_K_S.gguf create mode 100644 TheDrummer_Rocinante-X-12B-v1-Q5_K_L.gguf create mode 100644 TheDrummer_Rocinante-X-12B-v1-Q5_K_M.gguf create mode 100644 TheDrummer_Rocinante-X-12B-v1-Q5_K_S.gguf create mode 100644 TheDrummer_Rocinante-X-12B-v1-Q6_K.gguf create mode 100644 TheDrummer_Rocinante-X-12B-v1-Q6_K_L.gguf create mode 100644 TheDrummer_Rocinante-X-12B-v1-Q8_0.gguf create mode 100644 TheDrummer_Rocinante-X-12B-v1-bf16.gguf create mode 100644 TheDrummer_Rocinante-X-12B-v1-imatrix.gguf diff --git a/.gitattributes b/.gitattributes new file mode 100644 index 0000000..6cae62e --- /dev/null +++ b/.gitattributes @@ -0,0 +1,61 @@ +*.7z filter=lfs diff=lfs merge=lfs -text +*.arrow filter=lfs diff=lfs merge=lfs -text +*.bin filter=lfs diff=lfs merge=lfs -text +*.bz2 filter=lfs diff=lfs merge=lfs -text +*.ckpt 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 +*.mlmodel filter=lfs diff=lfs merge=lfs -text +*.model filter=lfs diff=lfs merge=lfs -text +*.msgpack filter=lfs diff=lfs merge=lfs -text +*.npy filter=lfs diff=lfs merge=lfs -text +*.npz 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 +*.pickle filter=lfs diff=lfs merge=lfs -text +*.pkl 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 +*.safetensors filter=lfs diff=lfs merge=lfs -text +saved_model/**/* filter=lfs diff=lfs merge=lfs -text +*.tar.* 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 +*.wasm filter=lfs diff=lfs merge=lfs -text +*.xz filter=lfs diff=lfs merge=lfs -text +*.zip filter=lfs diff=lfs merge=lfs -text +*.zst filter=lfs diff=lfs merge=lfs -text +*tfevents* filter=lfs diff=lfs merge=lfs -text +TheDrummer_Rocinante-X-12B-v1-Q8_0.gguf filter=lfs diff=lfs merge=lfs -text +TheDrummer_Rocinante-X-12B-v1-Q6_K_L.gguf filter=lfs diff=lfs merge=lfs -text +TheDrummer_Rocinante-X-12B-v1-Q6_K.gguf filter=lfs diff=lfs merge=lfs -text +TheDrummer_Rocinante-X-12B-v1-Q5_K_L.gguf filter=lfs diff=lfs merge=lfs -text +TheDrummer_Rocinante-X-12B-v1-Q5_K_M.gguf filter=lfs diff=lfs merge=lfs -text +TheDrummer_Rocinante-X-12B-v1-Q5_K_S.gguf filter=lfs diff=lfs merge=lfs -text +TheDrummer_Rocinante-X-12B-v1-Q4_K_L.gguf filter=lfs diff=lfs merge=lfs -text +TheDrummer_Rocinante-X-12B-v1-Q4_K_M.gguf filter=lfs diff=lfs merge=lfs -text +TheDrummer_Rocinante-X-12B-v1-Q4_K_S.gguf filter=lfs diff=lfs merge=lfs -text +TheDrummer_Rocinante-X-12B-v1-Q4_1.gguf filter=lfs diff=lfs merge=lfs -text +TheDrummer_Rocinante-X-12B-v1-Q4_0.gguf filter=lfs diff=lfs merge=lfs -text +TheDrummer_Rocinante-X-12B-v1-IQ4_NL.gguf filter=lfs diff=lfs merge=lfs -text +TheDrummer_Rocinante-X-12B-v1-IQ4_XS.gguf filter=lfs diff=lfs merge=lfs -text +TheDrummer_Rocinante-X-12B-v1-Q3_K_XL.gguf filter=lfs diff=lfs merge=lfs -text +TheDrummer_Rocinante-X-12B-v1-Q3_K_L.gguf filter=lfs diff=lfs merge=lfs -text +TheDrummer_Rocinante-X-12B-v1-Q3_K_M.gguf filter=lfs diff=lfs merge=lfs -text +TheDrummer_Rocinante-X-12B-v1-IQ3_M.gguf filter=lfs diff=lfs merge=lfs -text +TheDrummer_Rocinante-X-12B-v1-Q3_K_S.gguf filter=lfs diff=lfs merge=lfs -text +TheDrummer_Rocinante-X-12B-v1-IQ3_XS.gguf filter=lfs diff=lfs merge=lfs -text +TheDrummer_Rocinante-X-12B-v1-IQ3_XXS.gguf filter=lfs diff=lfs merge=lfs -text +TheDrummer_Rocinante-X-12B-v1-Q2_K_L.gguf filter=lfs diff=lfs merge=lfs -text +TheDrummer_Rocinante-X-12B-v1-Q2_K.gguf filter=lfs diff=lfs merge=lfs -text +TheDrummer_Rocinante-X-12B-v1-IQ2_M.gguf filter=lfs diff=lfs merge=lfs -text +TheDrummer_Rocinante-X-12B-v1-IQ2_S.gguf filter=lfs diff=lfs merge=lfs -text +TheDrummer_Rocinante-X-12B-v1-bf16.gguf filter=lfs diff=lfs merge=lfs -text +TheDrummer_Rocinante-X-12B-v1-imatrix.gguf filter=lfs diff=lfs merge=lfs -text diff --git a/README.md b/README.md new file mode 100644 index 0000000..6280912 --- /dev/null +++ b/README.md @@ -0,0 +1,176 @@ +--- +quantized_by: bartowski +pipeline_tag: text-generation +base_model: TheDrummer/Rocinante-X-12B-v1 +base_model_relation: quantized +--- + +## Llamacpp imatrix Quantizations of Rocinante-X-12B-v1 by TheDrummer + +Using llama.cpp release b7779 for quantization. + +Original model: https://huggingface.co/TheDrummer/Rocinante-X-12B-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 + +``` +[INST]{system_prompt} + +{prompt}[/INST] +``` + +## Download a file (not the whole branch) from below: + +| Filename | Quant type | File Size | Split | Description | +| -------- | ---------- | --------- | ----- | ----------- | +| [Rocinante-X-12B-v1-bf16.gguf](https://huggingface.co/bartowski/TheDrummer_Rocinante-X-12B-v1-GGUF/blob/main/TheDrummer_Rocinante-X-12B-v1-bf16.gguf) | bf16 | 24.50GB | false | Full BF16 weights. | +| [Rocinante-X-12B-v1-Q8_0.gguf](https://huggingface.co/bartowski/TheDrummer_Rocinante-X-12B-v1-GGUF/blob/main/TheDrummer_Rocinante-X-12B-v1-Q8_0.gguf) | Q8_0 | 13.02GB | false | Extremely high quality, generally unneeded but max available quant. | +| [Rocinante-X-12B-v1-Q6_K_L.gguf](https://huggingface.co/bartowski/TheDrummer_Rocinante-X-12B-v1-GGUF/blob/main/TheDrummer_Rocinante-X-12B-v1-Q6_K_L.gguf) | Q6_K_L | 10.38GB | false | Uses Q8_0 for embed and output weights. Very high quality, near perfect, *recommended*. | +| [Rocinante-X-12B-v1-Q6_K.gguf](https://huggingface.co/bartowski/TheDrummer_Rocinante-X-12B-v1-GGUF/blob/main/TheDrummer_Rocinante-X-12B-v1-Q6_K.gguf) | Q6_K | 10.06GB | false | Very high quality, near perfect, *recommended*. | +| [Rocinante-X-12B-v1-Q5_K_L.gguf](https://huggingface.co/bartowski/TheDrummer_Rocinante-X-12B-v1-GGUF/blob/main/TheDrummer_Rocinante-X-12B-v1-Q5_K_L.gguf) | Q5_K_L | 9.14GB | false | Uses Q8_0 for embed and output weights. High quality, *recommended*. | +| [Rocinante-X-12B-v1-Q5_K_M.gguf](https://huggingface.co/bartowski/TheDrummer_Rocinante-X-12B-v1-GGUF/blob/main/TheDrummer_Rocinante-X-12B-v1-Q5_K_M.gguf) | Q5_K_M | 8.73GB | false | High quality, *recommended*. | +| [Rocinante-X-12B-v1-Q5_K_S.gguf](https://huggingface.co/bartowski/TheDrummer_Rocinante-X-12B-v1-GGUF/blob/main/TheDrummer_Rocinante-X-12B-v1-Q5_K_S.gguf) | Q5_K_S | 8.52GB | false | High quality, *recommended*. | +| [Rocinante-X-12B-v1-Q4_K_L.gguf](https://huggingface.co/bartowski/TheDrummer_Rocinante-X-12B-v1-GGUF/blob/main/TheDrummer_Rocinante-X-12B-v1-Q4_K_L.gguf) | Q4_K_L | 7.98GB | false | Uses Q8_0 for embed and output weights. Good quality, *recommended*. | +| [Rocinante-X-12B-v1-Q4_1.gguf](https://huggingface.co/bartowski/TheDrummer_Rocinante-X-12B-v1-GGUF/blob/main/TheDrummer_Rocinante-X-12B-v1-Q4_1.gguf) | Q4_1 | 7.80GB | false | Legacy format, similar performance to Q4_K_S but with improved tokens/watt on Apple silicon. | +| [Rocinante-X-12B-v1-Q4_K_M.gguf](https://huggingface.co/bartowski/TheDrummer_Rocinante-X-12B-v1-GGUF/blob/main/TheDrummer_Rocinante-X-12B-v1-Q4_K_M.gguf) | Q4_K_M | 7.48GB | false | Good quality, default size for most use cases, *recommended*. | +| [Rocinante-X-12B-v1-Q3_K_XL.gguf](https://huggingface.co/bartowski/TheDrummer_Rocinante-X-12B-v1-GGUF/blob/main/TheDrummer_Rocinante-X-12B-v1-Q3_K_XL.gguf) | Q3_K_XL | 7.15GB | false | Uses Q8_0 for embed and output weights. Lower quality but usable, good for low RAM availability. | +| [Rocinante-X-12B-v1-Q4_K_S.gguf](https://huggingface.co/bartowski/TheDrummer_Rocinante-X-12B-v1-GGUF/blob/main/TheDrummer_Rocinante-X-12B-v1-Q4_K_S.gguf) | Q4_K_S | 7.12GB | false | Slightly lower quality with more space savings, *recommended*. | +| [Rocinante-X-12B-v1-IQ4_NL.gguf](https://huggingface.co/bartowski/TheDrummer_Rocinante-X-12B-v1-GGUF/blob/main/TheDrummer_Rocinante-X-12B-v1-IQ4_NL.gguf) | IQ4_NL | 7.10GB | false | Similar to IQ4_XS, but slightly larger. Offers online repacking for ARM CPU inference. | +| [Rocinante-X-12B-v1-Q4_0.gguf](https://huggingface.co/bartowski/TheDrummer_Rocinante-X-12B-v1-GGUF/blob/main/TheDrummer_Rocinante-X-12B-v1-Q4_0.gguf) | Q4_0 | 7.09GB | false | Legacy format, offers online repacking for ARM and AVX CPU inference. | +| [Rocinante-X-12B-v1-IQ4_XS.gguf](https://huggingface.co/bartowski/TheDrummer_Rocinante-X-12B-v1-GGUF/blob/main/TheDrummer_Rocinante-X-12B-v1-IQ4_XS.gguf) | IQ4_XS | 6.74GB | false | Decent quality, smaller than Q4_K_S with similar performance, *recommended*. | +| [Rocinante-X-12B-v1-Q3_K_L.gguf](https://huggingface.co/bartowski/TheDrummer_Rocinante-X-12B-v1-GGUF/blob/main/TheDrummer_Rocinante-X-12B-v1-Q3_K_L.gguf) | Q3_K_L | 6.56GB | false | Lower quality but usable, good for low RAM availability. | +| [Rocinante-X-12B-v1-Q3_K_M.gguf](https://huggingface.co/bartowski/TheDrummer_Rocinante-X-12B-v1-GGUF/blob/main/TheDrummer_Rocinante-X-12B-v1-Q3_K_M.gguf) | Q3_K_M | 6.08GB | false | Low quality. | +| [Rocinante-X-12B-v1-IQ3_M.gguf](https://huggingface.co/bartowski/TheDrummer_Rocinante-X-12B-v1-GGUF/blob/main/TheDrummer_Rocinante-X-12B-v1-IQ3_M.gguf) | IQ3_M | 5.72GB | false | Medium-low quality, new method with decent performance comparable to Q3_K_M. | +| [Rocinante-X-12B-v1-Q3_K_S.gguf](https://huggingface.co/bartowski/TheDrummer_Rocinante-X-12B-v1-GGUF/blob/main/TheDrummer_Rocinante-X-12B-v1-Q3_K_S.gguf) | Q3_K_S | 5.53GB | false | Low quality, not recommended. | +| [Rocinante-X-12B-v1-Q2_K_L.gguf](https://huggingface.co/bartowski/TheDrummer_Rocinante-X-12B-v1-GGUF/blob/main/TheDrummer_Rocinante-X-12B-v1-Q2_K_L.gguf) | Q2_K_L | 5.45GB | false | Uses Q8_0 for embed and output weights. Very low quality but surprisingly usable. | +| [Rocinante-X-12B-v1-IQ3_XS.gguf](https://huggingface.co/bartowski/TheDrummer_Rocinante-X-12B-v1-GGUF/blob/main/TheDrummer_Rocinante-X-12B-v1-IQ3_XS.gguf) | IQ3_XS | 5.31GB | false | Lower quality, new method with decent performance, slightly better than Q3_K_S. | +| [Rocinante-X-12B-v1-IQ3_XXS.gguf](https://huggingface.co/bartowski/TheDrummer_Rocinante-X-12B-v1-GGUF/blob/main/TheDrummer_Rocinante-X-12B-v1-IQ3_XXS.gguf) | IQ3_XXS | 4.95GB | false | Lower quality, new method with decent performance, comparable to Q3 quants. | +| [Rocinante-X-12B-v1-Q2_K.gguf](https://huggingface.co/bartowski/TheDrummer_Rocinante-X-12B-v1-GGUF/blob/main/TheDrummer_Rocinante-X-12B-v1-Q2_K.gguf) | Q2_K | 4.79GB | false | Very low quality but surprisingly usable. | +| [Rocinante-X-12B-v1-IQ2_M.gguf](https://huggingface.co/bartowski/TheDrummer_Rocinante-X-12B-v1-GGUF/blob/main/TheDrummer_Rocinante-X-12B-v1-IQ2_M.gguf) | IQ2_M | 4.44GB | false | Relatively low quality, uses SOTA techniques to be surprisingly usable. | +| [Rocinante-X-12B-v1-IQ2_S.gguf](https://huggingface.co/bartowski/TheDrummer_Rocinante-X-12B-v1-GGUF/blob/main/TheDrummer_Rocinante-X-12B-v1-IQ2_S.gguf) | IQ2_S | 4.14GB | 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/TheDrummer_Rocinante-X-12B-v1-GGUF --include "TheDrummer_Rocinante-X-12B-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_Rocinante-X-12B-v1-GGUF --include "TheDrummer_Rocinante-X-12B-v1-Q8_0/*" --local-dir ./ +``` + +You can either specify a new local-dir (TheDrummer_Rocinante-X-12B-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_Rocinante-X-12B-v1-IQ2_M.gguf b/TheDrummer_Rocinante-X-12B-v1-IQ2_M.gguf new file mode 100644 index 0000000..5b7f02b --- /dev/null +++ b/TheDrummer_Rocinante-X-12B-v1-IQ2_M.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c3a2de488c83a2d5e92170490f595e60e6686adb7091e14fd12ea8c31cd1d228 +size 4435026592 diff --git a/TheDrummer_Rocinante-X-12B-v1-IQ2_S.gguf b/TheDrummer_Rocinante-X-12B-v1-IQ2_S.gguf new file mode 100644 index 0000000..b5c2aac --- /dev/null +++ b/TheDrummer_Rocinante-X-12B-v1-IQ2_S.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a1350c097624cd15ad1b9465833f7d8d82b935f399af981bdd3d9c06f7b1f859 +size 4138476192 diff --git a/TheDrummer_Rocinante-X-12B-v1-IQ3_M.gguf b/TheDrummer_Rocinante-X-12B-v1-IQ3_M.gguf new file mode 100644 index 0000000..c4482aa --- /dev/null +++ b/TheDrummer_Rocinante-X-12B-v1-IQ3_M.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fe1630827bb189a359f6ac519a21746085c61a4a457bca083a9883289cf86322 +size 5722235552 diff --git a/TheDrummer_Rocinante-X-12B-v1-IQ3_XS.gguf b/TheDrummer_Rocinante-X-12B-v1-IQ3_XS.gguf new file mode 100644 index 0000000..f87c59f --- /dev/null +++ b/TheDrummer_Rocinante-X-12B-v1-IQ3_XS.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9862dcd9cbc1940042e6de1cde11a4312c95ac76d0d82f5506d4674fac0a611c +size 5306491552 diff --git a/TheDrummer_Rocinante-X-12B-v1-IQ3_XXS.gguf b/TheDrummer_Rocinante-X-12B-v1-IQ3_XXS.gguf new file mode 100644 index 0000000..c08e418 --- /dev/null +++ b/TheDrummer_Rocinante-X-12B-v1-IQ3_XXS.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:56f60553a641ffaca107429cdae203626389ed94268b2bfca41bf820966aaa33 +size 4945388192 diff --git a/TheDrummer_Rocinante-X-12B-v1-IQ4_NL.gguf b/TheDrummer_Rocinante-X-12B-v1-IQ4_NL.gguf new file mode 100644 index 0000000..a09e980 --- /dev/null +++ b/TheDrummer_Rocinante-X-12B-v1-IQ4_NL.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8b2f29c522cd1e9a9b4caa55ed45756ae8f16311c180aa564025d030500200e8 +size 7097918112 diff --git a/TheDrummer_Rocinante-X-12B-v1-IQ4_XS.gguf b/TheDrummer_Rocinante-X-12B-v1-IQ4_XS.gguf new file mode 100644 index 0000000..553b635 --- /dev/null +++ b/TheDrummer_Rocinante-X-12B-v1-IQ4_XS.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:43298ff2dced905224734a28ab0dbbc4801c1f66dbb72a170226142507804d29 +size 6742712992 diff --git a/TheDrummer_Rocinante-X-12B-v1-Q2_K.gguf b/TheDrummer_Rocinante-X-12B-v1-Q2_K.gguf new file mode 100644 index 0000000..6daba93 --- /dev/null +++ b/TheDrummer_Rocinante-X-12B-v1-Q2_K.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8f0c66f5a4e78b047a32dbcd9c50495a8c12bdb5e701accf9cb53ac0d197a28d +size 4791050912 diff --git a/TheDrummer_Rocinante-X-12B-v1-Q2_K_L.gguf b/TheDrummer_Rocinante-X-12B-v1-Q2_K_L.gguf new file mode 100644 index 0000000..7122d0f --- /dev/null +++ b/TheDrummer_Rocinante-X-12B-v1-Q2_K_L.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:62cdde6c1f1ce3051d7a18017be36b2e6fa8b14c6e6b67aa4296336dfda72025 +size 5446410912 diff --git a/TheDrummer_Rocinante-X-12B-v1-Q3_K_L.gguf b/TheDrummer_Rocinante-X-12B-v1-Q3_K_L.gguf new file mode 100644 index 0000000..18e8fa6 --- /dev/null +++ b/TheDrummer_Rocinante-X-12B-v1-Q3_K_L.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:49c28494d4bb6ae2ba0ca225270125c23d2246505c9ffd7e6bf50c375ab65778 +size 6561505952 diff --git a/TheDrummer_Rocinante-X-12B-v1-Q3_K_M.gguf b/TheDrummer_Rocinante-X-12B-v1-Q3_K_M.gguf new file mode 100644 index 0000000..652ceb4 --- /dev/null +++ b/TheDrummer_Rocinante-X-12B-v1-Q3_K_M.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bfb86d52acce0c892efe89f89d94a5dca73108a302c8f14f2d4dac41febb2b22 +size 6083093152 diff --git a/TheDrummer_Rocinante-X-12B-v1-Q3_K_S.gguf b/TheDrummer_Rocinante-X-12B-v1-Q3_K_S.gguf new file mode 100644 index 0000000..de15fbc --- /dev/null +++ b/TheDrummer_Rocinante-X-12B-v1-Q3_K_S.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9c65c3af862cf7e4b640063b2f3edb546b4bdf53daf423ef6f455cd2fe0ec956 +size 5534229152 diff --git a/TheDrummer_Rocinante-X-12B-v1-Q3_K_XL.gguf b/TheDrummer_Rocinante-X-12B-v1-Q3_K_XL.gguf new file mode 100644 index 0000000..c376752 --- /dev/null +++ b/TheDrummer_Rocinante-X-12B-v1-Q3_K_XL.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:301946387392763753b5a028ff6df7963f68ca4aaad21551b40af7909eea8571 +size 7148708512 diff --git a/TheDrummer_Rocinante-X-12B-v1-Q4_0.gguf b/TheDrummer_Rocinante-X-12B-v1-Q4_0.gguf new file mode 100644 index 0000000..27e480b --- /dev/null +++ b/TheDrummer_Rocinante-X-12B-v1-Q4_0.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:64b9208001561ddc9d794d836f5ee86c95174baa492aa7d93bababebfa13e299 +size 7094641312 diff --git a/TheDrummer_Rocinante-X-12B-v1-Q4_1.gguf b/TheDrummer_Rocinante-X-12B-v1-Q4_1.gguf new file mode 100644 index 0000000..1d537fd --- /dev/null +++ b/TheDrummer_Rocinante-X-12B-v1-Q4_1.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c3c04792701b7b4ffddcbe6b58c5c172a0fb88239c472bcff5986f64c3c52f69 +size 7795221152 diff --git a/TheDrummer_Rocinante-X-12B-v1-Q4_K_L.gguf b/TheDrummer_Rocinante-X-12B-v1-Q4_K_L.gguf new file mode 100644 index 0000000..d64d1fe --- /dev/null +++ b/TheDrummer_Rocinante-X-12B-v1-Q4_K_L.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8cb4564abd4ed071b26fba03a192a18724f98d6fc56ba96f6d0eeaeb6f503fbb +size 7975281312 diff --git a/TheDrummer_Rocinante-X-12B-v1-Q4_K_M.gguf b/TheDrummer_Rocinante-X-12B-v1-Q4_K_M.gguf new file mode 100644 index 0000000..be87b3b --- /dev/null +++ b/TheDrummer_Rocinante-X-12B-v1-Q4_K_M.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1ca25853a19d6b15f9a94f7e5bc86f4cc88bd0c9dea12a00ea55452b4208ade6 +size 7477207712 diff --git a/TheDrummer_Rocinante-X-12B-v1-Q4_K_S.gguf b/TheDrummer_Rocinante-X-12B-v1-Q4_K_S.gguf new file mode 100644 index 0000000..393a833 --- /dev/null +++ b/TheDrummer_Rocinante-X-12B-v1-Q4_K_S.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bfb2a88b3c4a5f7a0c6db0ff994909f7a2999342f12a97aaad696c23f11b1647 +size 7120200352 diff --git a/TheDrummer_Rocinante-X-12B-v1-Q5_K_L.gguf b/TheDrummer_Rocinante-X-12B-v1-Q5_K_L.gguf new file mode 100644 index 0000000..7d7bb6e --- /dev/null +++ b/TheDrummer_Rocinante-X-12B-v1-Q5_K_L.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8de520685db4c178bf187622e44cfc655321fb79c98cbe95c2ca132e5abf3506 +size 9141822112 diff --git a/TheDrummer_Rocinante-X-12B-v1-Q5_K_M.gguf b/TheDrummer_Rocinante-X-12B-v1-Q5_K_M.gguf new file mode 100644 index 0000000..278d858 --- /dev/null +++ b/TheDrummer_Rocinante-X-12B-v1-Q5_K_M.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:84a25f1d1c7eb00adf825702b57b8e0f57f32df0be6355e13fa7b03deebe98e2 +size 8727634592 diff --git a/TheDrummer_Rocinante-X-12B-v1-Q5_K_S.gguf b/TheDrummer_Rocinante-X-12B-v1-Q5_K_S.gguf new file mode 100644 index 0000000..be70eb2 --- /dev/null +++ b/TheDrummer_Rocinante-X-12B-v1-Q5_K_S.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:14738c094988fc60ea01d327cac7f7a150f2f634cc76f5dd22c9f0aa8d929f5e +size 8518738592 diff --git a/TheDrummer_Rocinante-X-12B-v1-Q6_K.gguf b/TheDrummer_Rocinante-X-12B-v1-Q6_K.gguf new file mode 100644 index 0000000..5498132 --- /dev/null +++ b/TheDrummer_Rocinante-X-12B-v1-Q6_K.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5c9ad1c2cd856c6c8ed424a3929afab91a5ac2b4f8e3e60d5761f8e1b6835215 +size 10056213152 diff --git a/TheDrummer_Rocinante-X-12B-v1-Q6_K_L.gguf b/TheDrummer_Rocinante-X-12B-v1-Q6_K_L.gguf new file mode 100644 index 0000000..dc2771c --- /dev/null +++ b/TheDrummer_Rocinante-X-12B-v1-Q6_K_L.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6e0a98b2b20aa54f8af0ae71ba133a491637c66ba98d4b618f087df5a0295467 +size 10381271712 diff --git a/TheDrummer_Rocinante-X-12B-v1-Q8_0.gguf b/TheDrummer_Rocinante-X-12B-v1-Q8_0.gguf new file mode 100644 index 0000000..0243133 --- /dev/null +++ b/TheDrummer_Rocinante-X-12B-v1-Q8_0.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0b948c0e7f51de220761dc7dc6e3f858530abbdb35562f6c2f31525c1537a686 +size 13022372512 diff --git a/TheDrummer_Rocinante-X-12B-v1-bf16.gguf b/TheDrummer_Rocinante-X-12B-v1-bf16.gguf new file mode 100644 index 0000000..1504a66 --- /dev/null +++ b/TheDrummer_Rocinante-X-12B-v1-bf16.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a8b7a970f5e016a6ddb151a10b59befde4d1c115a3b276a3dc5b71cc23cbecdc +size 24504279424 diff --git a/TheDrummer_Rocinante-X-12B-v1-imatrix.gguf b/TheDrummer_Rocinante-X-12B-v1-imatrix.gguf new file mode 100644 index 0000000..04a1cd0 --- /dev/null +++ b/TheDrummer_Rocinante-X-12B-v1-imatrix.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bbd1688d8a31b90b300b6614ea59cada4937dcd4e1a8b63f69ee27e0d570313d +size 7088224