commit 4e3cd5201b33841960dbfed95bc5ca6a48cd9fdd Author: ModelHub XC Date: Thu Jul 30 18:03:14 2026 +0800 初始化项目，由ModelHub XC社区提供模型 Model: bartowski/Menlo_Lucy-GGUF Source: Original Platform diff --git a/.gitattributes b/.gitattributes new file mode 100644 index 0000000..938df87 --- /dev/null +++ b/.gitattributes @@ -0,0 +1,49 @@ +*.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 + +Menlo_Lucy.imatrix filter=lfs diff=lfs merge=lfs -text \ No newline at end of file diff --git a/Menlo_Lucy-IQ3_M.gguf b/Menlo_Lucy-IQ3_M.gguf new file mode 100644 index 0000000..55cee40 --- /dev/null +++ b/Menlo_Lucy-IQ3_M.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:52aa49ff8a7b9dbebb04cc527b0d9faf12a642f2673e21eea7179d485c95605d +size 895662336 diff --git a/Menlo_Lucy-IQ3_XS.gguf b/Menlo_Lucy-IQ3_XS.gguf new file mode 100644 index 0000000..0ae12df --- /dev/null +++ b/Menlo_Lucy-IQ3_XS.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6fcd320e8ff8ee0d6df69f887954ee5c4193efe10acf3f14fa9c0ce776986b7f +size 834222336 diff --git a/Menlo_Lucy-IQ3_XXS.gguf b/Menlo_Lucy-IQ3_XXS.gguf new file mode 100644 index 0000000..c87d3e0 --- /dev/null +++ b/Menlo_Lucy-IQ3_XXS.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8d258325d003a6ddf6e7d4fa88d771ca2da0a1140505a13109966301f39c5550 +size 754360576 diff --git a/Menlo_Lucy-IQ4_NL.gguf b/Menlo_Lucy-IQ4_NL.gguf new file mode 100644 index 0000000..de7f7f1 --- /dev/null +++ b/Menlo_Lucy-IQ4_NL.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:58b8f21bbf77ba033cbdb5ad6913cfc6f4b07587c028bcfc61a6a56c13788fad +size 1054423296 diff --git a/Menlo_Lucy-IQ4_XS.gguf b/Menlo_Lucy-IQ4_XS.gguf new file mode 100644 index 0000000..ad11957 --- /dev/null +++ b/Menlo_Lucy-IQ4_XS.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:084c2eb1f1a57db53deed2966903088b24f3abd43cf752eb808d106e13067a95 +size 1010383104 diff --git a/Menlo_Lucy-Q2_K.gguf b/Menlo_Lucy-Q2_K.gguf new file mode 100644 index 0000000..9479b68 --- /dev/null +++ b/Menlo_Lucy-Q2_K.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:828e60220d03dfafd0cc535ea5da169b48d08c678f45f21cd7694488d51929cb +size 777795840 diff --git a/Menlo_Lucy-Q2_K_L.gguf b/Menlo_Lucy-Q2_K_L.gguf new file mode 100644 index 0000000..77071b2 --- /dev/null +++ b/Menlo_Lucy-Q2_K_L.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c7c5b5dbb913c526048f2626a65987f2d5002eef6eb895abc513d58e57e87c86 +size 853156096 diff --git a/Menlo_Lucy-Q3_K_L.gguf b/Menlo_Lucy-Q3_K_L.gguf new file mode 100644 index 0000000..90a4534 --- /dev/null +++ b/Menlo_Lucy-Q3_K_L.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ea25e375b9fc059c5862e33ea643e6436f78811f3c78d3f669a74077c10c84b3 +size 1003501824 diff --git a/Menlo_Lucy-Q3_K_M.gguf b/Menlo_Lucy-Q3_K_M.gguf new file mode 100644 index 0000000..0e91027 --- /dev/null +++ b/Menlo_Lucy-Q3_K_M.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ccbe16c64e5ea2be006b2cb392a187f574f67f6d30eaf490ea1d38cd7c320d40 +size 939538688 diff --git a/Menlo_Lucy-Q3_K_S.gguf b/Menlo_Lucy-Q3_K_S.gguf new file mode 100644 index 0000000..2952f28 --- /dev/null +++ b/Menlo_Lucy-Q3_K_S.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:713127d7fcce7c2ec100f7075947dc3aa7934b4f3f68b1e65e2f5e5b49e78e2e +size 867252480 diff --git a/Menlo_Lucy-Q3_K_XL.gguf b/Menlo_Lucy-Q3_K_XL.gguf new file mode 100644 index 0000000..010aaa3 --- /dev/null +++ b/Menlo_Lucy-Q3_K_XL.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a451a71c70c735fe92298942188655ea0df09570bff0b17a6b6b17f079dabfe4 +size 1078862080 diff --git a/Menlo_Lucy-Q4_0.gguf b/Menlo_Lucy-Q4_0.gguf new file mode 100644 index 0000000..9690d3f --- /dev/null +++ b/Menlo_Lucy-Q4_0.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9cca90a3e4e558f1b9f31a4eff9b0bbe4e3b5edd3e8c3bb2e94e674b33b82bbd +size 1056782592 diff --git a/Menlo_Lucy-Q4_1.gguf b/Menlo_Lucy-Q4_1.gguf new file mode 100644 index 0000000..3c00062 --- /dev/null +++ b/Menlo_Lucy-Q4_1.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:62f1f9e6de97668bf84cb03efdeb6688deda35e32c7f87c2e2886ea616eba1c4 +size 1142503680 diff --git a/Menlo_Lucy-Q4_K_L.gguf b/Menlo_Lucy-Q4_K_L.gguf new file mode 100644 index 0000000..87ef2a8 --- /dev/null +++ b/Menlo_Lucy-Q4_K_L.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d1e36862a0b3f8103b63982b9c85ab06c135334632d6c07c482267b41fed1871 +size 1182769408 diff --git a/Menlo_Lucy-Q4_K_M.gguf b/Menlo_Lucy-Q4_K_M.gguf new file mode 100644 index 0000000..dde0557 --- /dev/null +++ b/Menlo_Lucy-Q4_K_M.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1cb1682a9dbea9a1c8406721695f3faf6a212554d283585f2ec4608921f7c8b7 +size 1107409152 diff --git a/Menlo_Lucy-Q4_K_S.gguf b/Menlo_Lucy-Q4_K_S.gguf new file mode 100644 index 0000000..42a0bbf --- /dev/null +++ b/Menlo_Lucy-Q4_K_S.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5bfd50775505bf4973bb71d2ca5cb2055d5526d89957bc4271e4c978a7d3081d +size 1060190464 diff --git a/Menlo_Lucy-Q5_K_L.gguf b/Menlo_Lucy-Q5_K_L.gguf new file mode 100644 index 0000000..d595fe5 --- /dev/null +++ b/Menlo_Lucy-Q5_K_L.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b5dbb7d460fabade0749e1e117e141ea45be62b10fefa6e3c98f2585088333ec +size 1333240064 diff --git a/Menlo_Lucy-Q5_K_M.gguf b/Menlo_Lucy-Q5_K_M.gguf new file mode 100644 index 0000000..442d059 --- /dev/null +++ b/Menlo_Lucy-Q5_K_M.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a3b4ab6d49b3cfaba6c87b91baf38463cc6ea275859c09add1739f00182853ac +size 1257879808 diff --git a/Menlo_Lucy-Q5_K_S.gguf b/Menlo_Lucy-Q5_K_S.gguf new file mode 100644 index 0000000..2d1524a --- /dev/null +++ b/Menlo_Lucy-Q5_K_S.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ce413aff07e9dbc522028b70a1c8641d4d400b7adc8b360e3c87e997323d7be5 +size 1230584064 diff --git a/Menlo_Lucy-Q6_K.gguf b/Menlo_Lucy-Q6_K.gguf new file mode 100644 index 0000000..aca44ee --- /dev/null +++ b/Menlo_Lucy-Q6_K.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:10b44e2118c4411cd84dc6c6b0c737a53bc22e3fae1d006aab10765340a36744 +size 1417754880 diff --git a/Menlo_Lucy-Q6_K_L.gguf b/Menlo_Lucy-Q6_K_L.gguf new file mode 100644 index 0000000..3627767 --- /dev/null +++ b/Menlo_Lucy-Q6_K_L.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d63ab222dccbd0ebebd8a071b255ce6a64f6964dfe4f425532877dfe86f7e152 +size 1493115136 diff --git a/Menlo_Lucy-Q8_0.gguf b/Menlo_Lucy-Q8_0.gguf new file mode 100644 index 0000000..e2f4827 --- /dev/null +++ b/Menlo_Lucy-Q8_0.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0e888cd21367cfd7700e020faa85a6d546b0ca9359933b1e53dcaf92e79f14a1 +size 1834426624 diff --git a/Menlo_Lucy-bf16.gguf b/Menlo_Lucy-bf16.gguf new file mode 100644 index 0000000..2d5b6b1 --- /dev/null +++ b/Menlo_Lucy-bf16.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0d733d788d845589a29689e8e344e667fafb08ca12897598a916295066c54d04 +size 3447349248 diff --git a/Menlo_Lucy.imatrix b/Menlo_Lucy.imatrix new file mode 100644 index 0000000..a56eefe --- /dev/null +++ b/Menlo_Lucy.imatrix @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:26a01efa4201b2055b61daeb5fe871bfb294270a5af87f661e206f3e69cdd371 +size 2070886 diff --git a/README.md b/README.md new file mode 100644 index 0000000..d441efe --- /dev/null +++ b/README.md @@ -0,0 +1,171 @@ +--- +quantized_by: bartowski +pipeline_tag: text-generation +base_model: Menlo/Lucy +base_model_relation: quantized +--- + +## Llamacpp imatrix Quantizations of Lucy by Menlo + +Using llama.cpp release b5924 for quantization. + +Original model: https://huggingface.co/Menlo/Lucy + +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 + +No chat template specified so default is used. This may be incorrect, check original model card for details. + +``` +<|im_start|>system +{system_prompt}<|im_end|> +<|im_start|>user +{prompt}<|im_end|> +<|im_start|>assistant +``` + +## Download a file (not the whole branch) from below: + +| Filename | Quant type | File Size | Split | Description | +| -------- | ---------- | --------- | ----- | ----------- | +| [Lucy-bf16.gguf](https://huggingface.co/bartowski/Menlo_Lucy-GGUF/blob/main/Menlo_Lucy-bf16.gguf) | bf16 | 3.45GB | false | Full BF16 weights. | +| [Lucy-Q8_0.gguf](https://huggingface.co/bartowski/Menlo_Lucy-GGUF/blob/main/Menlo_Lucy-Q8_0.gguf) | Q8_0 | 1.83GB | false | Extremely high quality, generally unneeded but max available quant. | +| [Lucy-Q6_K_L.gguf](https://huggingface.co/bartowski/Menlo_Lucy-GGUF/blob/main/Menlo_Lucy-Q6_K_L.gguf) | Q6_K_L | 1.49GB | false | Uses Q8_0 for embed and output weights. Very high quality, near perfect, *recommended*. | +| [Lucy-Q6_K.gguf](https://huggingface.co/bartowski/Menlo_Lucy-GGUF/blob/main/Menlo_Lucy-Q6_K.gguf) | Q6_K | 1.42GB | false | Very high quality, near perfect, *recommended*. | +| [Lucy-Q5_K_L.gguf](https://huggingface.co/bartowski/Menlo_Lucy-GGUF/blob/main/Menlo_Lucy-Q5_K_L.gguf) | Q5_K_L | 1.33GB | false | Uses Q8_0 for embed and output weights. High quality, *recommended*. | +| [Lucy-Q5_K_M.gguf](https://huggingface.co/bartowski/Menlo_Lucy-GGUF/blob/main/Menlo_Lucy-Q5_K_M.gguf) | Q5_K_M | 1.26GB | false | High quality, *recommended*. | +| [Lucy-Q5_K_S.gguf](https://huggingface.co/bartowski/Menlo_Lucy-GGUF/blob/main/Menlo_Lucy-Q5_K_S.gguf) | Q5_K_S | 1.23GB | false | High quality, *recommended*. | +| [Lucy-Q4_K_L.gguf](https://huggingface.co/bartowski/Menlo_Lucy-GGUF/blob/main/Menlo_Lucy-Q4_K_L.gguf) | Q4_K_L | 1.18GB | false | Uses Q8_0 for embed and output weights. Good quality, *recommended*. | +| [Lucy-Q4_1.gguf](https://huggingface.co/bartowski/Menlo_Lucy-GGUF/blob/main/Menlo_Lucy-Q4_1.gguf) | Q4_1 | 1.14GB | false | Legacy format, similar performance to Q4_K_S but with improved tokens/watt on Apple silicon. | +| [Lucy-Q4_K_M.gguf](https://huggingface.co/bartowski/Menlo_Lucy-GGUF/blob/main/Menlo_Lucy-Q4_K_M.gguf) | Q4_K_M | 1.11GB | false | Good quality, default size for most use cases, *recommended*. | +| [Lucy-Q3_K_XL.gguf](https://huggingface.co/bartowski/Menlo_Lucy-GGUF/blob/main/Menlo_Lucy-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. | +| [Lucy-Q4_K_S.gguf](https://huggingface.co/bartowski/Menlo_Lucy-GGUF/blob/main/Menlo_Lucy-Q4_K_S.gguf) | Q4_K_S | 1.06GB | false | Slightly lower quality with more space savings, *recommended*. | +| [Lucy-Q4_0.gguf](https://huggingface.co/bartowski/Menlo_Lucy-GGUF/blob/main/Menlo_Lucy-Q4_0.gguf) | Q4_0 | 1.06GB | false | Legacy format, offers online repacking for ARM and AVX CPU inference. | +| [Lucy-IQ4_NL.gguf](https://huggingface.co/bartowski/Menlo_Lucy-GGUF/blob/main/Menlo_Lucy-IQ4_NL.gguf) | IQ4_NL | 1.05GB | false | Similar to IQ4_XS, but slightly larger. Offers online repacking for ARM CPU inference. | +| [Lucy-IQ4_XS.gguf](https://huggingface.co/bartowski/Menlo_Lucy-GGUF/blob/main/Menlo_Lucy-IQ4_XS.gguf) | IQ4_XS | 1.01GB | false | Decent quality, smaller than Q4_K_S with similar performance, *recommended*. | +| [Lucy-Q3_K_L.gguf](https://huggingface.co/bartowski/Menlo_Lucy-GGUF/blob/main/Menlo_Lucy-Q3_K_L.gguf) | Q3_K_L | 1.00GB | false | Lower quality but usable, good for low RAM availability. | +| [Lucy-Q3_K_M.gguf](https://huggingface.co/bartowski/Menlo_Lucy-GGUF/blob/main/Menlo_Lucy-Q3_K_M.gguf) | Q3_K_M | 0.94GB | false | Low quality. | +| [Lucy-IQ3_M.gguf](https://huggingface.co/bartowski/Menlo_Lucy-GGUF/blob/main/Menlo_Lucy-IQ3_M.gguf) | IQ3_M | 0.90GB | false | Medium-low quality, new method with decent performance comparable to Q3_K_M. | +| [Lucy-Q3_K_S.gguf](https://huggingface.co/bartowski/Menlo_Lucy-GGUF/blob/main/Menlo_Lucy-Q3_K_S.gguf) | Q3_K_S | 0.87GB | false | Low quality, not recommended. | +| [Lucy-Q2_K_L.gguf](https://huggingface.co/bartowski/Menlo_Lucy-GGUF/blob/main/Menlo_Lucy-Q2_K_L.gguf) | Q2_K_L | 0.85GB | false | Uses Q8_0 for embed and output weights. Very low quality but surprisingly usable. | +| [Lucy-IQ3_XS.gguf](https://huggingface.co/bartowski/Menlo_Lucy-GGUF/blob/main/Menlo_Lucy-IQ3_XS.gguf) | IQ3_XS | 0.83GB | false | Lower quality, new method with decent performance, slightly better than Q3_K_S. | +| [Lucy-Q2_K.gguf](https://huggingface.co/bartowski/Menlo_Lucy-GGUF/blob/main/Menlo_Lucy-Q2_K.gguf) | Q2_K | 0.78GB | false | Very low quality but surprisingly usable. | +| [Lucy-IQ3_XXS.gguf](https://huggingface.co/bartowski/Menlo_Lucy-GGUF/blob/main/Menlo_Lucy-IQ3_XXS.gguf) | IQ3_XXS | 0.75GB | false | Lower quality, new method with decent performance, comparable to Q3 quants. | + +## 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/Menlo_Lucy-GGUF --include "Menlo_Lucy-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/Menlo_Lucy-GGUF --include "Menlo_Lucy-Q8_0/*" --local-dir ./ +``` + +You can either specify a new local-dir (Menlo_Lucy-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