From 3199a5183c0b2cc025e91db605337bde602f259c Mon Sep 17 00:00:00 2001 From: ModelHub XC Date: Thu, 13 Aug 2026 09:41:13 +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/tencent_Hunyuan-1.8B-Instruct-GGUF Source: Original Platform --- .gitattributes | 47 ++++++ README.md | 166 +++++++++++++++++++++ configuration.json | 1 + tencent_Hunyuan-1.8B-Instruct-IQ2_M.gguf | 3 + tencent_Hunyuan-1.8B-Instruct-IQ3_M.gguf | 3 + tencent_Hunyuan-1.8B-Instruct-IQ3_XS.gguf | 3 + tencent_Hunyuan-1.8B-Instruct-IQ3_XXS.gguf | 3 + tencent_Hunyuan-1.8B-Instruct-IQ4_NL.gguf | 3 + tencent_Hunyuan-1.8B-Instruct-IQ4_XS.gguf | 3 + tencent_Hunyuan-1.8B-Instruct-Q2_K.gguf | 3 + tencent_Hunyuan-1.8B-Instruct-Q2_K_L.gguf | 3 + tencent_Hunyuan-1.8B-Instruct-Q3_K_L.gguf | 3 + tencent_Hunyuan-1.8B-Instruct-Q3_K_M.gguf | 3 + tencent_Hunyuan-1.8B-Instruct-Q3_K_S.gguf | 3 + tencent_Hunyuan-1.8B-Instruct-Q3_K_XL.gguf | 3 + tencent_Hunyuan-1.8B-Instruct-Q4_0.gguf | 3 + tencent_Hunyuan-1.8B-Instruct-Q4_1.gguf | 3 + tencent_Hunyuan-1.8B-Instruct-Q4_K_L.gguf | 3 + tencent_Hunyuan-1.8B-Instruct-Q4_K_M.gguf | 3 + tencent_Hunyuan-1.8B-Instruct-Q4_K_S.gguf | 3 + tencent_Hunyuan-1.8B-Instruct-Q5_K_L.gguf | 3 + tencent_Hunyuan-1.8B-Instruct-Q5_K_M.gguf | 3 + tencent_Hunyuan-1.8B-Instruct-Q5_K_S.gguf | 3 + tencent_Hunyuan-1.8B-Instruct-Q6_K.gguf | 3 + tencent_Hunyuan-1.8B-Instruct-Q6_K_L.gguf | 3 + tencent_Hunyuan-1.8B-Instruct-Q8_0.gguf | 3 + tencent_Hunyuan-1.8B-Instruct-bf16.gguf | 3 + tencent_Hunyuan-1.8B-Instruct-imatrix.gguf | 3 + 28 files changed, 289 insertions(+) create mode 100644 .gitattributes create mode 100644 README.md create mode 100644 configuration.json create mode 100644 tencent_Hunyuan-1.8B-Instruct-IQ2_M.gguf create mode 100644 tencent_Hunyuan-1.8B-Instruct-IQ3_M.gguf create mode 100644 tencent_Hunyuan-1.8B-Instruct-IQ3_XS.gguf create mode 100644 tencent_Hunyuan-1.8B-Instruct-IQ3_XXS.gguf create mode 100644 tencent_Hunyuan-1.8B-Instruct-IQ4_NL.gguf create mode 100644 tencent_Hunyuan-1.8B-Instruct-IQ4_XS.gguf create mode 100644 tencent_Hunyuan-1.8B-Instruct-Q2_K.gguf create mode 100644 tencent_Hunyuan-1.8B-Instruct-Q2_K_L.gguf create mode 100644 tencent_Hunyuan-1.8B-Instruct-Q3_K_L.gguf create mode 100644 tencent_Hunyuan-1.8B-Instruct-Q3_K_M.gguf create mode 100644 tencent_Hunyuan-1.8B-Instruct-Q3_K_S.gguf create mode 100644 tencent_Hunyuan-1.8B-Instruct-Q3_K_XL.gguf create mode 100644 tencent_Hunyuan-1.8B-Instruct-Q4_0.gguf create mode 100644 tencent_Hunyuan-1.8B-Instruct-Q4_1.gguf create mode 100644 tencent_Hunyuan-1.8B-Instruct-Q4_K_L.gguf create mode 100644 tencent_Hunyuan-1.8B-Instruct-Q4_K_M.gguf create mode 100644 tencent_Hunyuan-1.8B-Instruct-Q4_K_S.gguf create mode 100644 tencent_Hunyuan-1.8B-Instruct-Q5_K_L.gguf create mode 100644 tencent_Hunyuan-1.8B-Instruct-Q5_K_M.gguf create mode 100644 tencent_Hunyuan-1.8B-Instruct-Q5_K_S.gguf create mode 100644 tencent_Hunyuan-1.8B-Instruct-Q6_K.gguf create mode 100644 tencent_Hunyuan-1.8B-Instruct-Q6_K_L.gguf create mode 100644 tencent_Hunyuan-1.8B-Instruct-Q8_0.gguf create mode 100644 tencent_Hunyuan-1.8B-Instruct-bf16.gguf create mode 100644 tencent_Hunyuan-1.8B-Instruct-imatrix.gguf diff --git a/.gitattributes b/.gitattributes new file mode 100644 index 0000000..53d7257 --- /dev/null +++ b/.gitattributes @@ -0,0 +1,47 @@ +*.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 +*.gguf* 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 \ No newline at end of file diff --git a/README.md b/README.md new file mode 100644 index 0000000..bb3534b --- /dev/null +++ b/README.md @@ -0,0 +1,166 @@ +--- +quantized_by: bartowski +pipeline_tag: text-generation +base_model_relation: quantized +base_model: tencent/Hunyuan-1.8B-Instruct +--- + +## Llamacpp imatrix Quantizations of Hunyuan-1.8B-Instruct by tencent + +Using llama.cpp release b6076 for quantization. + +Original model: https://huggingface.co/tencent/Hunyuan-1.8B-Instruct + +All quants made using imatrix option with dataset from [here](https://gist.github.com/bartowski1182/eb213dccb3571f863da82e99418f81e8) + +Run them in [LM Studio](https://lmstudio.ai/) + +Run them directly with [llama.cpp](https://github.com/ggerganov/llama.cpp), or any other llama.cpp based project + +## Prompt format + +``` +<｜hy_begin▁of▁sentence｜>{system_prompt}<｜hy_place▁holder▁no▁3｜><｜hy_User｜>{prompt}<｜hy_Assistant｜><｜hy_place▁holder▁no▁2｜><｜hy_Assistant｜> +``` + +## Download a file (not the whole branch) from below: + +| Filename | Quant type | File Size | Split | Description | +| -------- | ---------- | --------- | ----- | ----------- | +| [Hunyuan-1.8B-Instruct-bf16.gguf](https://huggingface.co/bartowski/tencent_Hunyuan-1.8B-Instruct-GGUF/blob/main/tencent_Hunyuan-1.8B-Instruct-bf16.gguf) | bf16 | 3.59GB | false | Full BF16 weights. | +| [Hunyuan-1.8B-Instruct-Q8_0.gguf](https://huggingface.co/bartowski/tencent_Hunyuan-1.8B-Instruct-GGUF/blob/main/tencent_Hunyuan-1.8B-Instruct-Q8_0.gguf) | Q8_0 | 1.91GB | false | Extremely high quality, generally unneeded but max available quant. | +| [Hunyuan-1.8B-Instruct-Q6_K_L.gguf](https://huggingface.co/bartowski/tencent_Hunyuan-1.8B-Instruct-GGUF/blob/main/tencent_Hunyuan-1.8B-Instruct-Q6_K_L.gguf) | Q6_K_L | 1.53GB | false | Uses Q8_0 for embed and output weights. Very high quality, near perfect, *recommended*. | +| [Hunyuan-1.8B-Instruct-Q6_K.gguf](https://huggingface.co/bartowski/tencent_Hunyuan-1.8B-Instruct-GGUF/blob/main/tencent_Hunyuan-1.8B-Instruct-Q6_K.gguf) | Q6_K | 1.47GB | false | Very high quality, near perfect, *recommended*. | +| [Hunyuan-1.8B-Instruct-Q5_K_L.gguf](https://huggingface.co/bartowski/tencent_Hunyuan-1.8B-Instruct-GGUF/blob/main/tencent_Hunyuan-1.8B-Instruct-Q5_K_L.gguf) | Q5_K_L | 1.36GB | false | Uses Q8_0 for embed and output weights. High quality, *recommended*. | +| [Hunyuan-1.8B-Instruct-Q5_K_M.gguf](https://huggingface.co/bartowski/tencent_Hunyuan-1.8B-Instruct-GGUF/blob/main/tencent_Hunyuan-1.8B-Instruct-Q5_K_M.gguf) | Q5_K_M | 1.30GB | false | High quality, *recommended*. | +| [Hunyuan-1.8B-Instruct-Q5_K_S.gguf](https://huggingface.co/bartowski/tencent_Hunyuan-1.8B-Instruct-GGUF/blob/main/tencent_Hunyuan-1.8B-Instruct-Q5_K_S.gguf) | Q5_K_S | 1.27GB | false | High quality, *recommended*. | +| [Hunyuan-1.8B-Instruct-Q4_K_L.gguf](https://huggingface.co/bartowski/tencent_Hunyuan-1.8B-Instruct-GGUF/blob/main/tencent_Hunyuan-1.8B-Instruct-Q4_K_L.gguf) | Q4_K_L | 1.19GB | false | Uses Q8_0 for embed and output weights. Good quality, *recommended*. | +| [Hunyuan-1.8B-Instruct-Q4_1.gguf](https://huggingface.co/bartowski/tencent_Hunyuan-1.8B-Instruct-GGUF/blob/main/tencent_Hunyuan-1.8B-Instruct-Q4_1.gguf) | Q4_1 | 1.17GB | false | Legacy format, similar performance to Q4_K_S but with improved tokens/watt on Apple silicon. | +| [Hunyuan-1.8B-Instruct-Q4_K_M.gguf](https://huggingface.co/bartowski/tencent_Hunyuan-1.8B-Instruct-GGUF/blob/main/tencent_Hunyuan-1.8B-Instruct-Q4_K_M.gguf) | Q4_K_M | 1.13GB | false | Good quality, default size for most use cases, *recommended*. | +| [Hunyuan-1.8B-Instruct-Q4_K_S.gguf](https://huggingface.co/bartowski/tencent_Hunyuan-1.8B-Instruct-GGUF/blob/main/tencent_Hunyuan-1.8B-Instruct-Q4_K_S.gguf) | Q4_K_S | 1.08GB | false | Slightly lower quality with more space savings, *recommended*. | +| [Hunyuan-1.8B-Instruct-Q4_0.gguf](https://huggingface.co/bartowski/tencent_Hunyuan-1.8B-Instruct-GGUF/blob/main/tencent_Hunyuan-1.8B-Instruct-Q4_0.gguf) | Q4_0 | 1.08GB | false | Legacy format, offers online repacking for ARM and AVX CPU inference. | +| [Hunyuan-1.8B-Instruct-IQ4_NL.gguf](https://huggingface.co/bartowski/tencent_Hunyuan-1.8B-Instruct-GGUF/blob/main/tencent_Hunyuan-1.8B-Instruct-IQ4_NL.gguf) | IQ4_NL | 1.08GB | false | Similar to IQ4_XS, but slightly larger. Offers online repacking for ARM CPU inference. | +| [Hunyuan-1.8B-Instruct-Q3_K_XL.gguf](https://huggingface.co/bartowski/tencent_Hunyuan-1.8B-Instruct-GGUF/blob/main/tencent_Hunyuan-1.8B-Instruct-Q3_K_XL.gguf) | Q3_K_XL | 1.08GB | false | Uses Q8_0 for embed and output weights. Lower quality but usable, good for low RAM availability. | +| [Hunyuan-1.8B-Instruct-IQ4_XS.gguf](https://huggingface.co/bartowski/tencent_Hunyuan-1.8B-Instruct-GGUF/blob/main/tencent_Hunyuan-1.8B-Instruct-IQ4_XS.gguf) | IQ4_XS | 1.03GB | false | Decent quality, smaller than Q4_K_S with similar performance, *recommended*. | +| [Hunyuan-1.8B-Instruct-Q3_K_L.gguf](https://huggingface.co/bartowski/tencent_Hunyuan-1.8B-Instruct-GGUF/blob/main/tencent_Hunyuan-1.8B-Instruct-Q3_K_L.gguf) | Q3_K_L | 1.02GB | false | Lower quality but usable, good for low RAM availability. | +| [Hunyuan-1.8B-Instruct-Q3_K_M.gguf](https://huggingface.co/bartowski/tencent_Hunyuan-1.8B-Instruct-GGUF/blob/main/tencent_Hunyuan-1.8B-Instruct-Q3_K_M.gguf) | Q3_K_M | 0.95GB | false | Low quality. | +| [Hunyuan-1.8B-Instruct-IQ3_M.gguf](https://huggingface.co/bartowski/tencent_Hunyuan-1.8B-Instruct-GGUF/blob/main/tencent_Hunyuan-1.8B-Instruct-IQ3_M.gguf) | IQ3_M | 0.90GB | false | Medium-low quality, new method with decent performance comparable to Q3_K_M. | +| [Hunyuan-1.8B-Instruct-Q3_K_S.gguf](https://huggingface.co/bartowski/tencent_Hunyuan-1.8B-Instruct-GGUF/blob/main/tencent_Hunyuan-1.8B-Instruct-Q3_K_S.gguf) | Q3_K_S | 0.87GB | false | Low quality, not recommended. | +| [Hunyuan-1.8B-Instruct-IQ3_XS.gguf](https://huggingface.co/bartowski/tencent_Hunyuan-1.8B-Instruct-GGUF/blob/main/tencent_Hunyuan-1.8B-Instruct-IQ3_XS.gguf) | IQ3_XS | 0.84GB | false | Lower quality, new method with decent performance, slightly better than Q3_K_S. | +| [Hunyuan-1.8B-Instruct-Q2_K_L.gguf](https://huggingface.co/bartowski/tencent_Hunyuan-1.8B-Instruct-GGUF/blob/main/tencent_Hunyuan-1.8B-Instruct-Q2_K_L.gguf) | Q2_K_L | 0.84GB | false | Uses Q8_0 for embed and output weights. Very low quality but surprisingly usable. | +| [Hunyuan-1.8B-Instruct-Q2_K.gguf](https://huggingface.co/bartowski/tencent_Hunyuan-1.8B-Instruct-GGUF/blob/main/tencent_Hunyuan-1.8B-Instruct-Q2_K.gguf) | Q2_K | 0.78GB | false | Very low quality but surprisingly usable. | +| [Hunyuan-1.8B-Instruct-IQ3_XXS.gguf](https://huggingface.co/bartowski/tencent_Hunyuan-1.8B-Instruct-GGUF/blob/main/tencent_Hunyuan-1.8B-Instruct-IQ3_XXS.gguf) | IQ3_XXS | 0.77GB | false | Lower quality, new method with decent performance, comparable to Q3 quants. | +| [Hunyuan-1.8B-Instruct-IQ2_M.gguf](https://huggingface.co/bartowski/tencent_Hunyuan-1.8B-Instruct-GGUF/blob/main/tencent_Hunyuan-1.8B-Instruct-IQ2_M.gguf) | IQ2_M | 0.70GB | 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/tencent_Hunyuan-1.8B-Instruct-GGUF --include "tencent_Hunyuan-1.8B-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/tencent_Hunyuan-1.8B-Instruct-GGUF --include "tencent_Hunyuan-1.8B-Instruct-Q8_0/*" --local-dir ./ +``` + +You can either specify a new local-dir (tencent_Hunyuan-1.8B-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/ggerganov/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/ggerganov/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/ggerganov/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/ggerganov/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/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 diff --git a/tencent_Hunyuan-1.8B-Instruct-IQ2_M.gguf b/tencent_Hunyuan-1.8B-Instruct-IQ2_M.gguf new file mode 100644 index 0000000..9bb2860 --- /dev/null +++ b/tencent_Hunyuan-1.8B-Instruct-IQ2_M.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:32d6cde06fbf9f25705152a639bc5c22e4da03f392cdbafcdbea99bdb9f65c53 +size 698486688 diff --git a/tencent_Hunyuan-1.8B-Instruct-IQ3_M.gguf b/tencent_Hunyuan-1.8B-Instruct-IQ3_M.gguf new file mode 100644 index 0000000..6d312c3 --- /dev/null +++ b/tencent_Hunyuan-1.8B-Instruct-IQ3_M.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:537d02a6c8378cbafad65cec2f87651d071536e2a3ccf51d676adf9a71c093cb +size 900825280 diff --git a/tencent_Hunyuan-1.8B-Instruct-IQ3_XS.gguf b/tencent_Hunyuan-1.8B-Instruct-IQ3_XS.gguf new file mode 100644 index 0000000..936f6c3 --- /dev/null +++ b/tencent_Hunyuan-1.8B-Instruct-IQ3_XS.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:87f62086b9af852179f597d15b98aa7b4d43b702621bbbad1ecd66f5bd698c0a +size 840138944 diff --git a/tencent_Hunyuan-1.8B-Instruct-IQ3_XXS.gguf b/tencent_Hunyuan-1.8B-Instruct-IQ3_XXS.gguf new file mode 100644 index 0000000..43d9ee7 --- /dev/null +++ b/tencent_Hunyuan-1.8B-Instruct-IQ3_XXS.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2b4701242d9548e9e5e552e1a6a5965e3007dce846e41c3d655f73e0630a4583 +size 768479136 diff --git a/tencent_Hunyuan-1.8B-Instruct-IQ4_NL.gguf b/tencent_Hunyuan-1.8B-Instruct-IQ4_NL.gguf new file mode 100644 index 0000000..539502c --- /dev/null +++ b/tencent_Hunyuan-1.8B-Instruct-IQ4_NL.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:06ca69c9dc89f121073bf54fe48c2fdef152525a7974b9f4292f397407b8b5c7 +size 1081049280 diff --git a/tencent_Hunyuan-1.8B-Instruct-IQ4_XS.gguf b/tencent_Hunyuan-1.8B-Instruct-IQ4_XS.gguf new file mode 100644 index 0000000..1e5fc8a --- /dev/null +++ b/tencent_Hunyuan-1.8B-Instruct-IQ4_XS.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3f76334302985ab916be21b1450e794425a4ca741514cd5f55028ed0b2827271 +size 1033863360 diff --git a/tencent_Hunyuan-1.8B-Instruct-Q2_K.gguf b/tencent_Hunyuan-1.8B-Instruct-Q2_K.gguf new file mode 100644 index 0000000..520e9c5 --- /dev/null +++ b/tencent_Hunyuan-1.8B-Instruct-Q2_K.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9a6bcffe429feb90660b5da9fcf0ecf3ebae37fde24715ac025732a735ad9178 +size 777486528 diff --git a/tencent_Hunyuan-1.8B-Instruct-Q2_K_L.gguf b/tencent_Hunyuan-1.8B-Instruct-Q2_K_L.gguf new file mode 100644 index 0000000..67eda51 --- /dev/null +++ b/tencent_Hunyuan-1.8B-Instruct-Q2_K_L.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:88e7f945dfebfe9eabe9f135c986a2b79329d2298b375670d867e1e93a84cbc9 +size 837412256 diff --git a/tencent_Hunyuan-1.8B-Instruct-Q3_K_L.gguf b/tencent_Hunyuan-1.8B-Instruct-Q3_K_L.gguf new file mode 100644 index 0000000..86ff2bf --- /dev/null +++ b/tencent_Hunyuan-1.8B-Instruct-Q3_K_L.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8cd59623aa5bb0aa6b3e47c96b8ef9cae476c888b202ab38821ef45302bcc79d +size 1018921152 diff --git a/tencent_Hunyuan-1.8B-Instruct-Q3_K_M.gguf b/tencent_Hunyuan-1.8B-Instruct-Q3_K_M.gguf new file mode 100644 index 0000000..c00b090 --- /dev/null +++ b/tencent_Hunyuan-1.8B-Instruct-Q3_K_M.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a0b23b0528e1f53800469c1255bcaabeb7e8bb540237f8835ac3cd7357b327d4 +size 951025856 diff --git a/tencent_Hunyuan-1.8B-Instruct-Q3_K_S.gguf b/tencent_Hunyuan-1.8B-Instruct-Q3_K_S.gguf new file mode 100644 index 0000000..9802368 --- /dev/null +++ b/tencent_Hunyuan-1.8B-Instruct-Q3_K_S.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e225bd0a2b4406bdb684f8f1630eb2c60a93e4f8e39b088fd1e669935ad8641c +size 871858368 diff --git a/tencent_Hunyuan-1.8B-Instruct-Q3_K_XL.gguf b/tencent_Hunyuan-1.8B-Instruct-Q3_K_XL.gguf new file mode 100644 index 0000000..26ec4f2 --- /dev/null +++ b/tencent_Hunyuan-1.8B-Instruct-Q3_K_XL.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ddf002a31764891e77a5661a4cee124fa647911c63152697d9ff1cce2ec5dbcd +size 1078846880 diff --git a/tencent_Hunyuan-1.8B-Instruct-Q4_0.gguf b/tencent_Hunyuan-1.8B-Instruct-Q4_0.gguf new file mode 100644 index 0000000..3c4473a --- /dev/null +++ b/tencent_Hunyuan-1.8B-Instruct-Q4_0.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1fedacec05e3e4a0609bb992bacfb342e3877a4dd3d7c409e2577457c1069412 +size 1080000704 diff --git a/tencent_Hunyuan-1.8B-Instruct-Q4_1.gguf b/tencent_Hunyuan-1.8B-Instruct-Q4_1.gguf new file mode 100644 index 0000000..f6f66d1 --- /dev/null +++ b/tencent_Hunyuan-1.8B-Instruct-Q4_1.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:612695100cf6de2da41b5e9295e9e1a3b5347500f1765b227f8369d187781067 +size 1173323968 diff --git a/tencent_Hunyuan-1.8B-Instruct-Q4_K_L.gguf b/tencent_Hunyuan-1.8B-Instruct-Q4_K_L.gguf new file mode 100644 index 0000000..dfab21b --- /dev/null +++ b/tencent_Hunyuan-1.8B-Instruct-Q4_K_L.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:038db9592fadec7421683023f0b7441cb61cec1e2605ee7851429728948dcaa2 +size 1193010592 diff --git a/tencent_Hunyuan-1.8B-Instruct-Q4_K_M.gguf b/tencent_Hunyuan-1.8B-Instruct-Q4_K_M.gguf new file mode 100644 index 0000000..0035aaf --- /dev/null +++ b/tencent_Hunyuan-1.8B-Instruct-Q4_K_M.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:558ae86fc4251fd73cd54685319a83e7550997537e3ad41043ad5c0b5a2a26d4 +size 1133084864 diff --git a/tencent_Hunyuan-1.8B-Instruct-Q4_K_S.gguf b/tencent_Hunyuan-1.8B-Instruct-Q4_K_S.gguf new file mode 100644 index 0000000..1b4870e --- /dev/null +++ b/tencent_Hunyuan-1.8B-Instruct-Q4_K_S.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2a586cb1fb84f5e32c8136d65351cdf76d369ca61c42fc89bf10c66583b5bbad +size 1083670720 diff --git a/tencent_Hunyuan-1.8B-Instruct-Q5_K_L.gguf b/tencent_Hunyuan-1.8B-Instruct-Q5_K_L.gguf new file mode 100644 index 0000000..9838736 --- /dev/null +++ b/tencent_Hunyuan-1.8B-Instruct-Q5_K_L.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d1c85eccf8723fe2c5128f0a8e738dedc715af5ba41a60260a5ac59dcdef516f +size 1358685600 diff --git a/tencent_Hunyuan-1.8B-Instruct-Q5_K_M.gguf b/tencent_Hunyuan-1.8B-Instruct-Q5_K_M.gguf new file mode 100644 index 0000000..7962e80 --- /dev/null +++ b/tencent_Hunyuan-1.8B-Instruct-Q5_K_M.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f4a1fb2b91a8be0e20506dec6e7666776162731936a482763077787b8928d8e7 +size 1298759872 diff --git a/tencent_Hunyuan-1.8B-Instruct-Q5_K_S.gguf b/tencent_Hunyuan-1.8B-Instruct-Q5_K_S.gguf new file mode 100644 index 0000000..37e43bf --- /dev/null +++ b/tencent_Hunyuan-1.8B-Instruct-Q5_K_S.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8eb07c4f97c466dee2f89e20b92cd38890ba10be23175af20e6d37dc7d4e6201 +size 1269792960 diff --git a/tencent_Hunyuan-1.8B-Instruct-Q6_K.gguf b/tencent_Hunyuan-1.8B-Instruct-Q6_K.gguf new file mode 100644 index 0000000..831fac1 --- /dev/null +++ b/tencent_Hunyuan-1.8B-Instruct-Q6_K.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:db5faf346cd74411bd271c3c3c0555d6a44f5e05dcd2a4261a75c68b1b970bc1 +size 1474789568 diff --git a/tencent_Hunyuan-1.8B-Instruct-Q6_K_L.gguf b/tencent_Hunyuan-1.8B-Instruct-Q6_K_L.gguf new file mode 100644 index 0000000..f7b1991 --- /dev/null +++ b/tencent_Hunyuan-1.8B-Instruct-Q6_K_L.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:490699fc1d52c4c4f82ba6adeb7ab043cb8d094f33472eee0fb71a19aa57cbae +size 1534715296 diff --git a/tencent_Hunyuan-1.8B-Instruct-Q8_0.gguf b/tencent_Hunyuan-1.8B-Instruct-Q8_0.gguf new file mode 100644 index 0000000..d68c826 --- /dev/null +++ b/tencent_Hunyuan-1.8B-Instruct-Q8_0.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:93192de46d27cf2af1640313474589b9ff6434f26f59cb0a1d2ad336bf5a4124 +size 1908532640 diff --git a/tencent_Hunyuan-1.8B-Instruct-bf16.gguf b/tencent_Hunyuan-1.8B-Instruct-bf16.gguf new file mode 100644 index 0000000..a38b5bc --- /dev/null +++ b/tencent_Hunyuan-1.8B-Instruct-bf16.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:90bb812d9baca64f68d9d458797b7f7aa600ce90c73b5a82627ffd10eccd38d6 +size 3587537792 diff --git a/tencent_Hunyuan-1.8B-Instruct-imatrix.gguf b/tencent_Hunyuan-1.8B-Instruct-imatrix.gguf new file mode 100644 index 0000000..32bfdd1 --- /dev/null +++ b/tencent_Hunyuan-1.8B-Instruct-imatrix.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:10ff464f0eb599df2ac80e3c146406fea58142171d05b954e5d000052ae7ac79 +size 2393792