From 0617251aafa1bd0ebfe12a14382c870dc543f40b Mon Sep 17 00:00:00 2001 From: ModelHub XC Date: Mon, 13 Jul 2026 00:58:06 +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/allura-forge_Llama-3.3-8B-Instruct-GGUF Source: Original Platform --- .gitattributes | 73 ++++++++ README.md | 176 ++++++++++++++++++ allura-forge_Llama-3.3-8B-Instruct-IQ2_M.gguf | 3 + allura-forge_Llama-3.3-8B-Instruct-IQ3_M.gguf | 3 + ...ra-forge_Llama-3.3-8B-Instruct-IQ3_XS.gguf | 3 + ...a-forge_Llama-3.3-8B-Instruct-IQ3_XXS.gguf | 3 + ...ra-forge_Llama-3.3-8B-Instruct-IQ4_NL.gguf | 3 + ...ra-forge_Llama-3.3-8B-Instruct-IQ4_XS.gguf | 3 + allura-forge_Llama-3.3-8B-Instruct-Q2_K.gguf | 3 + ...ra-forge_Llama-3.3-8B-Instruct-Q2_K_L.gguf | 3 + ...ra-forge_Llama-3.3-8B-Instruct-Q3_K_L.gguf | 3 + ...ra-forge_Llama-3.3-8B-Instruct-Q3_K_M.gguf | 3 + ...ra-forge_Llama-3.3-8B-Instruct-Q3_K_S.gguf | 3 + ...a-forge_Llama-3.3-8B-Instruct-Q3_K_XL.gguf | 3 + allura-forge_Llama-3.3-8B-Instruct-Q4_0.gguf | 3 + allura-forge_Llama-3.3-8B-Instruct-Q4_1.gguf | 3 + ...ra-forge_Llama-3.3-8B-Instruct-Q4_K_L.gguf | 3 + ...ra-forge_Llama-3.3-8B-Instruct-Q4_K_M.gguf | 3 + ...ra-forge_Llama-3.3-8B-Instruct-Q4_K_S.gguf | 3 + ...ra-forge_Llama-3.3-8B-Instruct-Q5_K_L.gguf | 3 + ...ra-forge_Llama-3.3-8B-Instruct-Q5_K_M.gguf | 3 + ...ra-forge_Llama-3.3-8B-Instruct-Q5_K_S.gguf | 3 + allura-forge_Llama-3.3-8B-Instruct-Q6_K.gguf | 3 + ...ra-forge_Llama-3.3-8B-Instruct-Q6_K_L.gguf | 3 + allura-forge_Llama-3.3-8B-Instruct-Q8_0.gguf | 3 + allura-forge_Llama-3.3-8B-Instruct-bf16.gguf | 3 + ...a-forge_Llama-3.3-8B-Instruct-imatrix.gguf | 3 + configuration.json | 1 + 28 files changed, 325 insertions(+) create mode 100644 .gitattributes create mode 100644 README.md create mode 100644 allura-forge_Llama-3.3-8B-Instruct-IQ2_M.gguf create mode 100644 allura-forge_Llama-3.3-8B-Instruct-IQ3_M.gguf create mode 100644 allura-forge_Llama-3.3-8B-Instruct-IQ3_XS.gguf create mode 100644 allura-forge_Llama-3.3-8B-Instruct-IQ3_XXS.gguf create mode 100644 allura-forge_Llama-3.3-8B-Instruct-IQ4_NL.gguf create mode 100644 allura-forge_Llama-3.3-8B-Instruct-IQ4_XS.gguf create mode 100644 allura-forge_Llama-3.3-8B-Instruct-Q2_K.gguf create mode 100644 allura-forge_Llama-3.3-8B-Instruct-Q2_K_L.gguf create mode 100644 allura-forge_Llama-3.3-8B-Instruct-Q3_K_L.gguf create mode 100644 allura-forge_Llama-3.3-8B-Instruct-Q3_K_M.gguf create mode 100644 allura-forge_Llama-3.3-8B-Instruct-Q3_K_S.gguf create mode 100644 allura-forge_Llama-3.3-8B-Instruct-Q3_K_XL.gguf create mode 100644 allura-forge_Llama-3.3-8B-Instruct-Q4_0.gguf create mode 100644 allura-forge_Llama-3.3-8B-Instruct-Q4_1.gguf create mode 100644 allura-forge_Llama-3.3-8B-Instruct-Q4_K_L.gguf create mode 100644 allura-forge_Llama-3.3-8B-Instruct-Q4_K_M.gguf create mode 100644 allura-forge_Llama-3.3-8B-Instruct-Q4_K_S.gguf create mode 100644 allura-forge_Llama-3.3-8B-Instruct-Q5_K_L.gguf create mode 100644 allura-forge_Llama-3.3-8B-Instruct-Q5_K_M.gguf create mode 100644 allura-forge_Llama-3.3-8B-Instruct-Q5_K_S.gguf create mode 100644 allura-forge_Llama-3.3-8B-Instruct-Q6_K.gguf create mode 100644 allura-forge_Llama-3.3-8B-Instruct-Q6_K_L.gguf create mode 100644 allura-forge_Llama-3.3-8B-Instruct-Q8_0.gguf create mode 100644 allura-forge_Llama-3.3-8B-Instruct-bf16.gguf create mode 100644 allura-forge_Llama-3.3-8B-Instruct-imatrix.gguf create mode 100644 configuration.json diff --git a/.gitattributes b/.gitattributes new file mode 100644 index 0000000..0289723 --- /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 + +allura-forge_Llama-3.3-8B-Instruct-IQ3_XS.gguf filter=lfs diff=lfs merge=lfs -text +allura-forge_Llama-3.3-8B-Instruct-bf16.gguf filter=lfs diff=lfs merge=lfs -text +allura-forge_Llama-3.3-8B-Instruct-Q3_K_XL.gguf filter=lfs diff=lfs merge=lfs -text +allura-forge_Llama-3.3-8B-Instruct-Q2_K.gguf filter=lfs diff=lfs merge=lfs -text +allura-forge_Llama-3.3-8B-Instruct-Q4_0.gguf filter=lfs diff=lfs merge=lfs -text +allura-forge_Llama-3.3-8B-Instruct-IQ3_XXS.gguf filter=lfs diff=lfs merge=lfs -text +allura-forge_Llama-3.3-8B-Instruct-Q3_K_S.gguf filter=lfs diff=lfs merge=lfs -text +allura-forge_Llama-3.3-8B-Instruct-Q3_K_L.gguf filter=lfs diff=lfs merge=lfs -text +allura-forge_Llama-3.3-8B-Instruct-IQ2_M.gguf filter=lfs diff=lfs merge=lfs -text +allura-forge_Llama-3.3-8B-Instruct-Q3_K_M.gguf filter=lfs diff=lfs merge=lfs -text +allura-forge_Llama-3.3-8B-Instruct-Q6_K.gguf filter=lfs diff=lfs merge=lfs -text +allura-forge_Llama-3.3-8B-Instruct-Q4_K_M.gguf filter=lfs diff=lfs merge=lfs -text +allura-forge_Llama-3.3-8B-Instruct-Q5_K_S.gguf filter=lfs diff=lfs merge=lfs -text +allura-forge_Llama-3.3-8B-Instruct-Q4_1.gguf filter=lfs diff=lfs merge=lfs -text +allura-forge_Llama-3.3-8B-Instruct-Q6_K_L.gguf filter=lfs diff=lfs merge=lfs -text +allura-forge_Llama-3.3-8B-Instruct-imatrix.gguf filter=lfs diff=lfs merge=lfs -text +allura-forge_Llama-3.3-8B-Instruct-Q5_K_M.gguf filter=lfs diff=lfs merge=lfs -text +allura-forge_Llama-3.3-8B-Instruct-Q4_K_L.gguf filter=lfs diff=lfs merge=lfs -text +allura-forge_Llama-3.3-8B-Instruct-Q4_K_S.gguf filter=lfs diff=lfs merge=lfs -text +allura-forge_Llama-3.3-8B-Instruct-IQ4_NL.gguf filter=lfs diff=lfs merge=lfs -text +allura-forge_Llama-3.3-8B-Instruct-IQ3_M.gguf filter=lfs diff=lfs merge=lfs -text +allura-forge_Llama-3.3-8B-Instruct-IQ4_XS.gguf filter=lfs diff=lfs merge=lfs -text +allura-forge_Llama-3.3-8B-Instruct-Q2_K_L.gguf filter=lfs diff=lfs merge=lfs -text +allura-forge_Llama-3.3-8B-Instruct-Q8_0.gguf filter=lfs diff=lfs merge=lfs -text +allura-forge_Llama-3.3-8B-Instruct-Q5_K_L.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..aa9cbb2 --- /dev/null +++ b/README.md @@ -0,0 +1,176 @@ +--- +quantized_by: bartowski +pipeline_tag: text-generation +base_model: allura-forge/Llama-3.3-8B-Instruct +license: llama3.3 +base_model_relation: quantized +--- + +## Llamacpp imatrix Quantizations of Llama-3.3-8B-Instruct by allura-forge + +Using llama.cpp release b7545 for quantization. + +Original model: https://huggingface.co/allura-forge/Llama-3.3-8B-Instruct + +All quants made using imatrix option with dataset from [here](https://gist.github.com/bartowski1182/eb213dccb3571f863da82e99418f81e8) combined with a subset of combined_all_small.parquet from Ed Addario [here](https://huggingface.co/datasets/eaddario/imatrix-calibration/blob/main/combined_all_small.parquet) + +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) + +## Prompt format + +``` +<|begin_of_text|><|start_header_id|>system<|end_header_id|> + +{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 | +| -------- | ---------- | --------- | ----- | ----------- | +| [Llama-3.3-8B-Instruct-bf16.gguf](https://huggingface.co/bartowski/allura-forge_Llama-3.3-8B-Instruct-GGUF/blob/main/allura-forge_Llama-3.3-8B-Instruct-bf16.gguf) | bf16 | 16.07GB | false | Full BF16 weights. | +| [Llama-3.3-8B-Instruct-Q8_0.gguf](https://huggingface.co/bartowski/allura-forge_Llama-3.3-8B-Instruct-GGUF/blob/main/allura-forge_Llama-3.3-8B-Instruct-Q8_0.gguf) | Q8_0 | 8.54GB | false | Extremely high quality, generally unneeded but max available quant. | +| [Llama-3.3-8B-Instruct-Q6_K_L.gguf](https://huggingface.co/bartowski/allura-forge_Llama-3.3-8B-Instruct-GGUF/blob/main/allura-forge_Llama-3.3-8B-Instruct-Q6_K_L.gguf) | Q6_K_L | 6.85GB | false | Uses Q8_0 for embed and output weights. Very high quality, near perfect, *recommended*. | +| [Llama-3.3-8B-Instruct-Q6_K.gguf](https://huggingface.co/bartowski/allura-forge_Llama-3.3-8B-Instruct-GGUF/blob/main/allura-forge_Llama-3.3-8B-Instruct-Q6_K.gguf) | Q6_K | 6.60GB | false | Very high quality, near perfect, *recommended*. | +| [Llama-3.3-8B-Instruct-Q5_K_L.gguf](https://huggingface.co/bartowski/allura-forge_Llama-3.3-8B-Instruct-GGUF/blob/main/allura-forge_Llama-3.3-8B-Instruct-Q5_K_L.gguf) | Q5_K_L | 6.06GB | false | Uses Q8_0 for embed and output weights. High quality, *recommended*. | +| [Llama-3.3-8B-Instruct-Q5_K_M.gguf](https://huggingface.co/bartowski/allura-forge_Llama-3.3-8B-Instruct-GGUF/blob/main/allura-forge_Llama-3.3-8B-Instruct-Q5_K_M.gguf) | Q5_K_M | 5.73GB | false | High quality, *recommended*. | +| [Llama-3.3-8B-Instruct-Q5_K_S.gguf](https://huggingface.co/bartowski/allura-forge_Llama-3.3-8B-Instruct-GGUF/blob/main/allura-forge_Llama-3.3-8B-Instruct-Q5_K_S.gguf) | Q5_K_S | 5.60GB | false | High quality, *recommended*. | +| [Llama-3.3-8B-Instruct-Q4_K_L.gguf](https://huggingface.co/bartowski/allura-forge_Llama-3.3-8B-Instruct-GGUF/blob/main/allura-forge_Llama-3.3-8B-Instruct-Q4_K_L.gguf) | Q4_K_L | 5.31GB | false | Uses Q8_0 for embed and output weights. Good quality, *recommended*. | +| [Llama-3.3-8B-Instruct-Q4_1.gguf](https://huggingface.co/bartowski/allura-forge_Llama-3.3-8B-Instruct-GGUF/blob/main/allura-forge_Llama-3.3-8B-Instruct-Q4_1.gguf) | Q4_1 | 5.13GB | false | Legacy format, similar performance to Q4_K_S but with improved tokens/watt on Apple silicon. | +| [Llama-3.3-8B-Instruct-Q4_K_M.gguf](https://huggingface.co/bartowski/allura-forge_Llama-3.3-8B-Instruct-GGUF/blob/main/allura-forge_Llama-3.3-8B-Instruct-Q4_K_M.gguf) | Q4_K_M | 4.92GB | false | Good quality, default size for most use cases, *recommended*. | +| [Llama-3.3-8B-Instruct-Q3_K_XL.gguf](https://huggingface.co/bartowski/allura-forge_Llama-3.3-8B-Instruct-GGUF/blob/main/allura-forge_Llama-3.3-8B-Instruct-Q3_K_XL.gguf) | Q3_K_XL | 4.78GB | false | Uses Q8_0 for embed and output weights. Lower quality but usable, good for low RAM availability. | +| [Llama-3.3-8B-Instruct-Q4_K_S.gguf](https://huggingface.co/bartowski/allura-forge_Llama-3.3-8B-Instruct-GGUF/blob/main/allura-forge_Llama-3.3-8B-Instruct-Q4_K_S.gguf) | Q4_K_S | 4.69GB | false | Slightly lower quality with more space savings, *recommended*. | +| [Llama-3.3-8B-Instruct-Q4_0.gguf](https://huggingface.co/bartowski/allura-forge_Llama-3.3-8B-Instruct-GGUF/blob/main/allura-forge_Llama-3.3-8B-Instruct-Q4_0.gguf) | Q4_0 | 4.68GB | false | Legacy format, offers online repacking for ARM and AVX CPU inference. | +| [Llama-3.3-8B-Instruct-IQ4_NL.gguf](https://huggingface.co/bartowski/allura-forge_Llama-3.3-8B-Instruct-GGUF/blob/main/allura-forge_Llama-3.3-8B-Instruct-IQ4_NL.gguf) | IQ4_NL | 4.68GB | false | Similar to IQ4_XS, but slightly larger. Offers online repacking for ARM CPU inference. | +| [Llama-3.3-8B-Instruct-IQ4_XS.gguf](https://huggingface.co/bartowski/allura-forge_Llama-3.3-8B-Instruct-GGUF/blob/main/allura-forge_Llama-3.3-8B-Instruct-IQ4_XS.gguf) | IQ4_XS | 4.45GB | false | Decent quality, smaller than Q4_K_S with similar performance, *recommended*. | +| [Llama-3.3-8B-Instruct-Q3_K_L.gguf](https://huggingface.co/bartowski/allura-forge_Llama-3.3-8B-Instruct-GGUF/blob/main/allura-forge_Llama-3.3-8B-Instruct-Q3_K_L.gguf) | Q3_K_L | 4.32GB | false | Lower quality but usable, good for low RAM availability. | +| [Llama-3.3-8B-Instruct-Q3_K_M.gguf](https://huggingface.co/bartowski/allura-forge_Llama-3.3-8B-Instruct-GGUF/blob/main/allura-forge_Llama-3.3-8B-Instruct-Q3_K_M.gguf) | Q3_K_M | 4.02GB | false | Low quality. | +| [Llama-3.3-8B-Instruct-IQ3_M.gguf](https://huggingface.co/bartowski/allura-forge_Llama-3.3-8B-Instruct-GGUF/blob/main/allura-forge_Llama-3.3-8B-Instruct-IQ3_M.gguf) | IQ3_M | 3.78GB | false | Medium-low quality, new method with decent performance comparable to Q3_K_M. | +| [Llama-3.3-8B-Instruct-Q2_K_L.gguf](https://huggingface.co/bartowski/allura-forge_Llama-3.3-8B-Instruct-GGUF/blob/main/allura-forge_Llama-3.3-8B-Instruct-Q2_K_L.gguf) | Q2_K_L | 3.69GB | false | Uses Q8_0 for embed and output weights. Very low quality but surprisingly usable. | +| [Llama-3.3-8B-Instruct-Q3_K_S.gguf](https://huggingface.co/bartowski/allura-forge_Llama-3.3-8B-Instruct-GGUF/blob/main/allura-forge_Llama-3.3-8B-Instruct-Q3_K_S.gguf) | Q3_K_S | 3.66GB | false | Low quality, not recommended. | +| [Llama-3.3-8B-Instruct-IQ3_XS.gguf](https://huggingface.co/bartowski/allura-forge_Llama-3.3-8B-Instruct-GGUF/blob/main/allura-forge_Llama-3.3-8B-Instruct-IQ3_XS.gguf) | IQ3_XS | 3.52GB | false | Lower quality, new method with decent performance, slightly better than Q3_K_S. | +| [Llama-3.3-8B-Instruct-IQ3_XXS.gguf](https://huggingface.co/bartowski/allura-forge_Llama-3.3-8B-Instruct-GGUF/blob/main/allura-forge_Llama-3.3-8B-Instruct-IQ3_XXS.gguf) | IQ3_XXS | 3.27GB | false | Lower quality, new method with decent performance, comparable to Q3 quants. | +| [Llama-3.3-8B-Instruct-Q2_K.gguf](https://huggingface.co/bartowski/allura-forge_Llama-3.3-8B-Instruct-GGUF/blob/main/allura-forge_Llama-3.3-8B-Instruct-Q2_K.gguf) | Q2_K | 3.18GB | false | Very low quality but surprisingly usable. | +| [Llama-3.3-8B-Instruct-IQ2_M.gguf](https://huggingface.co/bartowski/allura-forge_Llama-3.3-8B-Instruct-GGUF/blob/main/allura-forge_Llama-3.3-8B-Instruct-IQ2_M.gguf) | IQ2_M | 2.95GB | 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/allura-forge_Llama-3.3-8B-Instruct-GGUF --include "allura-forge_Llama-3.3-8B-Instruct-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/allura-forge_Llama-3.3-8B-Instruct-GGUF --include "allura-forge_Llama-3.3-8B-Instruct-Q8_0/*" --local-dir ./ +``` + +You can either specify a new local-dir (allura-forge_Llama-3.3-8B-Instruct-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/allura-forge_Llama-3.3-8B-Instruct-IQ2_M.gguf b/allura-forge_Llama-3.3-8B-Instruct-IQ2_M.gguf new file mode 100644 index 0000000..c2dda82 --- /dev/null +++ b/allura-forge_Llama-3.3-8B-Instruct-IQ2_M.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f2bb8d7722272af496e38c1804bf6f239d8e32e0051406dc2b7513e2024fb686 +size 2948281344 diff --git a/allura-forge_Llama-3.3-8B-Instruct-IQ3_M.gguf b/allura-forge_Llama-3.3-8B-Instruct-IQ3_M.gguf new file mode 100644 index 0000000..186eb33 --- /dev/null +++ b/allura-forge_Llama-3.3-8B-Instruct-IQ3_M.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:39b72ab072eaba4a8f4f6430ce67ffab5849ea2260be8fe19e075ae34700e9ee +size 3784823808 diff --git a/allura-forge_Llama-3.3-8B-Instruct-IQ3_XS.gguf b/allura-forge_Llama-3.3-8B-Instruct-IQ3_XS.gguf new file mode 100644 index 0000000..bed5a7b --- /dev/null +++ b/allura-forge_Llama-3.3-8B-Instruct-IQ3_XS.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9f8028d27c3d1e68ca8c92c16c56f1a7eb03de9449216af76c035ff5734a85fa +size 3518747648 diff --git a/allura-forge_Llama-3.3-8B-Instruct-IQ3_XXS.gguf b/allura-forge_Llama-3.3-8B-Instruct-IQ3_XXS.gguf new file mode 100644 index 0000000..ff96413 --- /dev/null +++ b/allura-forge_Llama-3.3-8B-Instruct-IQ3_XXS.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:64e09ab80201b1f2d6da3be5a58c9fa8e64e185a491046c593e158aae529aaca +size 3274912768 diff --git a/allura-forge_Llama-3.3-8B-Instruct-IQ4_NL.gguf b/allura-forge_Llama-3.3-8B-Instruct-IQ4_NL.gguf new file mode 100644 index 0000000..99c5859 --- /dev/null +++ b/allura-forge_Llama-3.3-8B-Instruct-IQ4_NL.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b26b5d51c67b432b1e186ae2df815864ccfd179339cd6ef55f94d9b608fd049b +size 4677989376 diff --git a/allura-forge_Llama-3.3-8B-Instruct-IQ4_XS.gguf b/allura-forge_Llama-3.3-8B-Instruct-IQ4_XS.gguf new file mode 100644 index 0000000..b4e7d29 --- /dev/null +++ b/allura-forge_Llama-3.3-8B-Instruct-IQ4_XS.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8fb9b69008381fa90e35105bdfb8de6a1b57176af659520d3522ad7c138dbfd4 +size 4447663104 diff --git a/allura-forge_Llama-3.3-8B-Instruct-Q2_K.gguf b/allura-forge_Llama-3.3-8B-Instruct-Q2_K.gguf new file mode 100644 index 0000000..cd2c21d --- /dev/null +++ b/allura-forge_Llama-3.3-8B-Instruct-Q2_K.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:068a857667d0790b94c19129fbd490beb108ab27c3f11ce45b25b145aafd4f18 +size 3179131904 diff --git a/allura-forge_Llama-3.3-8B-Instruct-Q2_K_L.gguf b/allura-forge_Llama-3.3-8B-Instruct-Q2_K_L.gguf new file mode 100644 index 0000000..0448881 --- /dev/null +++ b/allura-forge_Llama-3.3-8B-Instruct-Q2_K_L.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e77cb6fd3fb666183b919a04809498f985b302ad577e748125fb0b8abb3f4b37 +size 3692155904 diff --git a/allura-forge_Llama-3.3-8B-Instruct-Q3_K_L.gguf b/allura-forge_Llama-3.3-8B-Instruct-Q3_K_L.gguf new file mode 100644 index 0000000..8676631 --- /dev/null +++ b/allura-forge_Llama-3.3-8B-Instruct-Q3_K_L.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d8be3779d9e6078b9998197e49f2c47f54d9c1e63c6839284fe06a28e7c16d5e +size 4321956864 diff --git a/allura-forge_Llama-3.3-8B-Instruct-Q3_K_M.gguf b/allura-forge_Llama-3.3-8B-Instruct-Q3_K_M.gguf new file mode 100644 index 0000000..e836578 --- /dev/null +++ b/allura-forge_Llama-3.3-8B-Instruct-Q3_K_M.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:521743be97ff2f94dc198f736cef9654331f584856f121b4fc931e2e4236c74c +size 4018918400 diff --git a/allura-forge_Llama-3.3-8B-Instruct-Q3_K_S.gguf b/allura-forge_Llama-3.3-8B-Instruct-Q3_K_S.gguf new file mode 100644 index 0000000..4c49ab3 --- /dev/null +++ b/allura-forge_Llama-3.3-8B-Instruct-Q3_K_S.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d2a6899852c7a671490a087175b5892f6790252f8b108d46f815512cc25fbdc0 +size 3664499712 diff --git a/allura-forge_Llama-3.3-8B-Instruct-Q3_K_XL.gguf b/allura-forge_Llama-3.3-8B-Instruct-Q3_K_XL.gguf new file mode 100644 index 0000000..19d1c8e --- /dev/null +++ b/allura-forge_Llama-3.3-8B-Instruct-Q3_K_XL.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1824335eee4e278e884adf361eae41abfc0f9e8d11fd355ad717103bf8f0ae03 +size 4781626368 diff --git a/allura-forge_Llama-3.3-8B-Instruct-Q4_0.gguf b/allura-forge_Llama-3.3-8B-Instruct-Q4_0.gguf new file mode 100644 index 0000000..268395b --- /dev/null +++ b/allura-forge_Llama-3.3-8B-Instruct-Q4_0.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5627874237fe617c2799ad64c4e5bf930f43f82eea09f7b6c13f17e87d016634 +size 4675892224 diff --git a/allura-forge_Llama-3.3-8B-Instruct-Q4_1.gguf b/allura-forge_Llama-3.3-8B-Instruct-Q4_1.gguf new file mode 100644 index 0000000..69cd89b --- /dev/null +++ b/allura-forge_Llama-3.3-8B-Instruct-Q4_1.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a00b3e8fc06e82181628765b5d69624015da1378b251b3866d6f733b1ccbfc1a +size 5130253312 diff --git a/allura-forge_Llama-3.3-8B-Instruct-Q4_K_L.gguf b/allura-forge_Llama-3.3-8B-Instruct-Q4_K_L.gguf new file mode 100644 index 0000000..60df9cc --- /dev/null +++ b/allura-forge_Llama-3.3-8B-Instruct-Q4_K_L.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e038d8254ade812eaee42fef262dbbf64f4f7e75f2dcd56440b0cf7e17862eea +size 5310632960 diff --git a/allura-forge_Llama-3.3-8B-Instruct-Q4_K_M.gguf b/allura-forge_Llama-3.3-8B-Instruct-Q4_K_M.gguf new file mode 100644 index 0000000..904f7f0 --- /dev/null +++ b/allura-forge_Llama-3.3-8B-Instruct-Q4_K_M.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8d7d3a45a884e6a4e7fc7d7de4024f987b78b0546a8e70c045a6046f777138c9 +size 4920734720 diff --git a/allura-forge_Llama-3.3-8B-Instruct-Q4_K_S.gguf b/allura-forge_Llama-3.3-8B-Instruct-Q4_K_S.gguf new file mode 100644 index 0000000..57e2413 --- /dev/null +++ b/allura-forge_Llama-3.3-8B-Instruct-Q4_K_S.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3d12c648f2bac6a8f75736bf06400d53295a6dba47f7e171a28e05a7e055ab87 +size 4692669440 diff --git a/allura-forge_Llama-3.3-8B-Instruct-Q5_K_L.gguf b/allura-forge_Llama-3.3-8B-Instruct-Q5_K_L.gguf new file mode 100644 index 0000000..963827f --- /dev/null +++ b/allura-forge_Llama-3.3-8B-Instruct-Q5_K_L.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f1349994042ecbc4a9086d6cb16704ac18a52301148e509e44f0c9b767b8074c +size 6057219072 diff --git a/allura-forge_Llama-3.3-8B-Instruct-Q5_K_M.gguf b/allura-forge_Llama-3.3-8B-Instruct-Q5_K_M.gguf new file mode 100644 index 0000000..a28534b --- /dev/null +++ b/allura-forge_Llama-3.3-8B-Instruct-Q5_K_M.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5065c7d3c02836ad58d78f84b11c535c486ae6fd50d9b42b954aea504b918f57 +size 5732987904 diff --git a/allura-forge_Llama-3.3-8B-Instruct-Q5_K_S.gguf b/allura-forge_Llama-3.3-8B-Instruct-Q5_K_S.gguf new file mode 100644 index 0000000..36126e3 --- /dev/null +++ b/allura-forge_Llama-3.3-8B-Instruct-Q5_K_S.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ac884d68305045f042da262aad411314af5d3ea84360edb1bfedd60f7b6164b6 +size 5599294464 diff --git a/allura-forge_Llama-3.3-8B-Instruct-Q6_K.gguf b/allura-forge_Llama-3.3-8B-Instruct-Q6_K.gguf new file mode 100644 index 0000000..3b64e55 --- /dev/null +++ b/allura-forge_Llama-3.3-8B-Instruct-Q6_K.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1a0ce04a85c57dca8c64d4d5017ad1d271774f1a3fad844dcf1cf75356a5b86b +size 6596006912 diff --git a/allura-forge_Llama-3.3-8B-Instruct-Q6_K_L.gguf b/allura-forge_Llama-3.3-8B-Instruct-Q6_K_L.gguf new file mode 100644 index 0000000..01b75a6 --- /dev/null +++ b/allura-forge_Llama-3.3-8B-Instruct-Q6_K_L.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8704a8ccec903ff085c21e9e9681bbe330acaccff5b556f9d4b00a530bdacaff +size 6850466816 diff --git a/allura-forge_Llama-3.3-8B-Instruct-Q8_0.gguf b/allura-forge_Llama-3.3-8B-Instruct-Q8_0.gguf new file mode 100644 index 0000000..f3ee34b --- /dev/null +++ b/allura-forge_Llama-3.3-8B-Instruct-Q8_0.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b0a1f6f011782e5e5d1ae1da7f97a8e23a3c306def218738d3a9a6ebd276b0af +size 8540771328 diff --git a/allura-forge_Llama-3.3-8B-Instruct-bf16.gguf b/allura-forge_Llama-3.3-8B-Instruct-bf16.gguf new file mode 100644 index 0000000..bceb08e --- /dev/null +++ b/allura-forge_Llama-3.3-8B-Instruct-bf16.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1bf4a17beaa49a9d06d81c3c231a7f922e35baec0ad3d798d8dddd75f5e9c904 +size 16068891360 diff --git a/allura-forge_Llama-3.3-8B-Instruct-imatrix.gguf b/allura-forge_Llama-3.3-8B-Instruct-imatrix.gguf new file mode 100644 index 0000000..6a3c566 --- /dev/null +++ b/allura-forge_Llama-3.3-8B-Instruct-imatrix.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b4adc5cdb988ea8bfb14ebe67b3f1cbe95451d73be711c49028a4d6db98c9aa2 +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