commit f6f92713c79aef5830eaa15f0effc195026ef16d Author: ModelHub XC Date: Tue Jul 21 18:09:07 2026 +0800 初始化项目，由ModelHub XC社区提供模型 Model: bartowski/Menlo_Jan-nano-128k-GGUF Source: Original Platform 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/Menlo_Jan-nano-128k-IQ2_M.gguf b/Menlo_Jan-nano-128k-IQ2_M.gguf new file mode 100644 index 0000000..bbc9e97 --- /dev/null +++ b/Menlo_Jan-nano-128k-IQ2_M.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1f24a3964012a33256680592c90949702c2e1dc253e08510a13cee15b46b2a6f +size 1512984032 diff --git a/Menlo_Jan-nano-128k-IQ3_M.gguf b/Menlo_Jan-nano-128k-IQ3_M.gguf new file mode 100644 index 0000000..1088a5e --- /dev/null +++ b/Menlo_Jan-nano-128k-IQ3_M.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7739fe3c2c100a5ff735bf775fe31049942d44669a30b3d6b0b4ef95f6543841 +size 1962896352 diff --git a/Menlo_Jan-nano-128k-IQ3_XS.gguf b/Menlo_Jan-nano-128k-IQ3_XS.gguf new file mode 100644 index 0000000..6b657b5 --- /dev/null +++ b/Menlo_Jan-nano-128k-IQ3_XS.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:eddd5fa11742fb30e704d38e9f68b1a90afca4634977f6a0fa5a5c42597b5101 +size 1814375392 diff --git a/Menlo_Jan-nano-128k-IQ3_XXS.gguf b/Menlo_Jan-nano-128k-IQ3_XXS.gguf new file mode 100644 index 0000000..7fbd753 --- /dev/null +++ b/Menlo_Jan-nano-128k-IQ3_XXS.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ed12593f49635f007edf4a50a42282f3e3c7a46a317d1df1d40a0373131eff55 +size 1670188512 diff --git a/Menlo_Jan-nano-128k-IQ4_NL.gguf b/Menlo_Jan-nano-128k-IQ4_NL.gguf new file mode 100644 index 0000000..4d56a44 --- /dev/null +++ b/Menlo_Jan-nano-128k-IQ4_NL.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f64ae21fcc9290d092d71d746029f49a2927418c8899647faa188a92361153d4 +size 2381343712 diff --git a/Menlo_Jan-nano-128k-IQ4_XS.gguf b/Menlo_Jan-nano-128k-IQ4_XS.gguf new file mode 100644 index 0000000..a51e1d4 --- /dev/null +++ b/Menlo_Jan-nano-128k-IQ4_XS.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:abfaf0d819fbb1f19089e2bf37dd514f2a8ed6ac0c1ad18f09b6c2e4febba2df +size 2270751712 diff --git a/Menlo_Jan-nano-128k-Q2_K.gguf b/Menlo_Jan-nano-128k-Q2_K.gguf new file mode 100644 index 0000000..d1892e5 --- /dev/null +++ b/Menlo_Jan-nano-128k-Q2_K.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:835ba94aeda5d4ba86e8cbc3893cafb65feae218dbb909dfc4fe6986af7e98c6 +size 1669499872 diff --git a/Menlo_Jan-nano-128k-Q2_K_L.gguf b/Menlo_Jan-nano-128k-Q2_K_L.gguf new file mode 100644 index 0000000..e556cd2 --- /dev/null +++ b/Menlo_Jan-nano-128k-Q2_K_L.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:33460ffa7cdb64f8d55640a32cfa0ccc15a4c0bbeaae639303375fe1d2083c0a +size 1763700192 diff --git a/Menlo_Jan-nano-128k-Q3_K_L.gguf b/Menlo_Jan-nano-128k-Q3_K_L.gguf new file mode 100644 index 0000000..2f77a1b --- /dev/null +++ b/Menlo_Jan-nano-128k-Q3_K_L.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b0e25432f0dfcdf85585dd4f05968d98551b023c07bdb8a25c4f09b639b335b4 +size 2239785952 diff --git a/Menlo_Jan-nano-128k-Q3_K_M.gguf b/Menlo_Jan-nano-128k-Q3_K_M.gguf new file mode 100644 index 0000000..6c40c6c --- /dev/null +++ b/Menlo_Jan-nano-128k-Q3_K_M.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:398ab8135a724b651d0a773501996eef0235300062cfbd3c2c4ea58de1c809c5 +size 2075618272 diff --git a/Menlo_Jan-nano-128k-Q3_K_S.gguf b/Menlo_Jan-nano-128k-Q3_K_S.gguf new file mode 100644 index 0000000..c873d47 --- /dev/null +++ b/Menlo_Jan-nano-128k-Q3_K_S.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:191a5e88e97f1efb3562269a1815809285ca11b064a8ac5e33a097271ef022f9 +size 1886997472 diff --git a/Menlo_Jan-nano-128k-Q3_K_XL.gguf b/Menlo_Jan-nano-128k-Q3_K_XL.gguf new file mode 100644 index 0000000..ce05704 --- /dev/null +++ b/Menlo_Jan-nano-128k-Q3_K_XL.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0682b7663fed04c1944286ca87186a919229377a27c87fc222a43b3d2d11c216 +size 2333986272 diff --git a/Menlo_Jan-nano-128k-Q4_0.gguf b/Menlo_Jan-nano-128k-Q4_0.gguf new file mode 100644 index 0000000..6826a84 --- /dev/null +++ b/Menlo_Jan-nano-128k-Q4_0.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0312132b321eb0a877c4e97360c6cc5ae97809d66cc96b2b784f7d9310f02390 +size 2375773152 diff --git a/Menlo_Jan-nano-128k-Q4_1.gguf b/Menlo_Jan-nano-128k-Q4_1.gguf new file mode 100644 index 0000000..87903e0 --- /dev/null +++ b/Menlo_Jan-nano-128k-Q4_1.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e974e743064022a611a79000fba40b8daf9b36cdcb867ec81c8bf81bd97fcc27 +size 2596629472 diff --git a/Menlo_Jan-nano-128k-Q4_K_L.gguf b/Menlo_Jan-nano-128k-Q4_K_L.gguf new file mode 100644 index 0000000..b014fa6 --- /dev/null +++ b/Menlo_Jan-nano-128k-Q4_K_L.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:df6a39577b8bff1323632f530a1f4370af55d364d9099f7c1820f86bb5050192 +size 2591481312 diff --git a/Menlo_Jan-nano-128k-Q4_K_M.gguf b/Menlo_Jan-nano-128k-Q4_K_M.gguf new file mode 100644 index 0000000..9a7c999 --- /dev/null +++ b/Menlo_Jan-nano-128k-Q4_K_M.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a864031a138288da427ca176afd61d7fe2b03fd19a84a656b2691aa1f7a12921 +size 2497280992 diff --git a/Menlo_Jan-nano-128k-Q4_K_S.gguf b/Menlo_Jan-nano-128k-Q4_K_S.gguf new file mode 100644 index 0000000..96f5a9f --- /dev/null +++ b/Menlo_Jan-nano-128k-Q4_K_S.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:01cd8f503d0d46030f3c120263d5808347aaba061d15d9674d94f4a1b22db20b +size 2383309792 diff --git a/Menlo_Jan-nano-128k-Q5_K_L.gguf b/Menlo_Jan-nano-128k-Q5_K_L.gguf new file mode 100644 index 0000000..f49676f --- /dev/null +++ b/Menlo_Jan-nano-128k-Q5_K_L.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e629f3196d8101d640f8cdaf9e98e7e1821b0c2b2cfe0a008cf79efd03da3f8b +size 2983714272 diff --git a/Menlo_Jan-nano-128k-Q5_K_M.gguf b/Menlo_Jan-nano-128k-Q5_K_M.gguf new file mode 100644 index 0000000..10993aa --- /dev/null +++ b/Menlo_Jan-nano-128k-Q5_K_M.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fa3e63a544b3553c92d2a82f1d64668edd1e567cbc432d714813719f258de801 +size 2889513952 diff --git a/Menlo_Jan-nano-128k-Q5_K_S.gguf b/Menlo_Jan-nano-128k-Q5_K_S.gguf new file mode 100644 index 0000000..e62d469 --- /dev/null +++ b/Menlo_Jan-nano-128k-Q5_K_S.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ab76dca5074e943b6b3eacce495d68107f487f0f4a03892e97e9dccc3710f756 +size 2823711712 diff --git a/Menlo_Jan-nano-128k-Q6_K.gguf b/Menlo_Jan-nano-128k-Q6_K.gguf new file mode 100644 index 0000000..6e01298 --- /dev/null +++ b/Menlo_Jan-nano-128k-Q6_K.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e2021d7318806fd693f14180d4525bb451d1aebe85822c295a9a684f52d6b8a0 +size 3306261472 diff --git a/Menlo_Jan-nano-128k-Q6_K_L.gguf b/Menlo_Jan-nano-128k-Q6_K_L.gguf new file mode 100644 index 0000000..bdb3d46 --- /dev/null +++ b/Menlo_Jan-nano-128k-Q6_K_L.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c04449403a4580b481599d81323c880e9e6851a7a76e7e76b4b78f1647af97a9 +size 3400461792 diff --git a/Menlo_Jan-nano-128k-Q8_0.gguf b/Menlo_Jan-nano-128k-Q8_0.gguf new file mode 100644 index 0000000..ccd6f4d --- /dev/null +++ b/Menlo_Jan-nano-128k-Q8_0.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2a454ce13866be8e3617ab9d28eea3f3c1fcc6d360189c3d4c8c6f73039f4c90 +size 4280405472 diff --git a/Menlo_Jan-nano-128k-bf16.gguf b/Menlo_Jan-nano-128k-bf16.gguf new file mode 100644 index 0000000..7e98cad --- /dev/null +++ b/Menlo_Jan-nano-128k-bf16.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:214ffd1408482c5587d73b7e684f2ec3f9daa9c9e3de8bc86cce2760ad0af723 +size 8051285216 diff --git a/Menlo_Jan-nano-128k.imatrix b/Menlo_Jan-nano-128k.imatrix new file mode 100644 index 0000000..bbb8ddd Binary files /dev/null and b/Menlo_Jan-nano-128k.imatrix differ diff --git a/README.md b/README.md new file mode 100644 index 0000000..f88a959 --- /dev/null +++ b/README.md @@ -0,0 +1,176 @@ +--- +quantized_by: bartowski +pipeline_tag: text-generation +base_model: Menlo/Jan-nano-128k +base_model_relation: quantized +license: apache-2.0 +language: +- en +--- + +## Llamacpp imatrix Quantizations of Jan-nano-128k by Menlo + +Using llama.cpp release b5697 for quantization. + +Original model: https://huggingface.co/Menlo/Jan-nano-128k + +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 + +``` +<|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 | +| -------- | ---------- | --------- | ----- | ----------- | +| [Jan-nano-128k-bf16.gguf](https://huggingface.co/bartowski/Menlo_Jan-nano-128k-GGUF/blob/main/Menlo_Jan-nano-128k-bf16.gguf) | bf16 | 8.05GB | false | Full BF16 weights. | +| [Jan-nano-128k-Q8_0.gguf](https://huggingface.co/bartowski/Menlo_Jan-nano-128k-GGUF/blob/main/Menlo_Jan-nano-128k-Q8_0.gguf) | Q8_0 | 4.28GB | false | Extremely high quality, generally unneeded but max available quant. | +| [Jan-nano-128k-Q6_K_L.gguf](https://huggingface.co/bartowski/Menlo_Jan-nano-128k-GGUF/blob/main/Menlo_Jan-nano-128k-Q6_K_L.gguf) | Q6_K_L | 3.40GB | false | Uses Q8_0 for embed and output weights. Very high quality, near perfect, *recommended*. | +| [Jan-nano-128k-Q6_K.gguf](https://huggingface.co/bartowski/Menlo_Jan-nano-128k-GGUF/blob/main/Menlo_Jan-nano-128k-Q6_K.gguf) | Q6_K | 3.31GB | false | Very high quality, near perfect, *recommended*. | +| [Jan-nano-128k-Q5_K_L.gguf](https://huggingface.co/bartowski/Menlo_Jan-nano-128k-GGUF/blob/main/Menlo_Jan-nano-128k-Q5_K_L.gguf) | Q5_K_L | 2.98GB | false | Uses Q8_0 for embed and output weights. High quality, *recommended*. | +| [Jan-nano-128k-Q5_K_M.gguf](https://huggingface.co/bartowski/Menlo_Jan-nano-128k-GGUF/blob/main/Menlo_Jan-nano-128k-Q5_K_M.gguf) | Q5_K_M | 2.89GB | false | High quality, *recommended*. | +| [Jan-nano-128k-Q5_K_S.gguf](https://huggingface.co/bartowski/Menlo_Jan-nano-128k-GGUF/blob/main/Menlo_Jan-nano-128k-Q5_K_S.gguf) | Q5_K_S | 2.82GB | false | High quality, *recommended*. | +| [Jan-nano-128k-Q4_1.gguf](https://huggingface.co/bartowski/Menlo_Jan-nano-128k-GGUF/blob/main/Menlo_Jan-nano-128k-Q4_1.gguf) | Q4_1 | 2.60GB | false | Legacy format, similar performance to Q4_K_S but with improved tokens/watt on Apple silicon. | +| [Jan-nano-128k-Q4_K_L.gguf](https://huggingface.co/bartowski/Menlo_Jan-nano-128k-GGUF/blob/main/Menlo_Jan-nano-128k-Q4_K_L.gguf) | Q4_K_L | 2.59GB | false | Uses Q8_0 for embed and output weights. Good quality, *recommended*. | +| [Jan-nano-128k-Q4_K_M.gguf](https://huggingface.co/bartowski/Menlo_Jan-nano-128k-GGUF/blob/main/Menlo_Jan-nano-128k-Q4_K_M.gguf) | Q4_K_M | 2.50GB | false | Good quality, default size for most use cases, *recommended*. | +| [Jan-nano-128k-Q4_K_S.gguf](https://huggingface.co/bartowski/Menlo_Jan-nano-128k-GGUF/blob/main/Menlo_Jan-nano-128k-Q4_K_S.gguf) | Q4_K_S | 2.38GB | false | Slightly lower quality with more space savings, *recommended*. | +| [Jan-nano-128k-Q4_0.gguf](https://huggingface.co/bartowski/Menlo_Jan-nano-128k-GGUF/blob/main/Menlo_Jan-nano-128k-Q4_0.gguf) | Q4_0 | 2.38GB | false | Legacy format, offers online repacking for ARM and AVX CPU inference. | +| [Jan-nano-128k-IQ4_NL.gguf](https://huggingface.co/bartowski/Menlo_Jan-nano-128k-GGUF/blob/main/Menlo_Jan-nano-128k-IQ4_NL.gguf) | IQ4_NL | 2.38GB | false | Similar to IQ4_XS, but slightly larger. Offers online repacking for ARM CPU inference. | +| [Jan-nano-128k-Q3_K_XL.gguf](https://huggingface.co/bartowski/Menlo_Jan-nano-128k-GGUF/blob/main/Menlo_Jan-nano-128k-Q3_K_XL.gguf) | Q3_K_XL | 2.33GB | false | Uses Q8_0 for embed and output weights. Lower quality but usable, good for low RAM availability. | +| [Jan-nano-128k-IQ4_XS.gguf](https://huggingface.co/bartowski/Menlo_Jan-nano-128k-GGUF/blob/main/Menlo_Jan-nano-128k-IQ4_XS.gguf) | IQ4_XS | 2.27GB | false | Decent quality, smaller than Q4_K_S with similar performance, *recommended*. | +| [Jan-nano-128k-Q3_K_L.gguf](https://huggingface.co/bartowski/Menlo_Jan-nano-128k-GGUF/blob/main/Menlo_Jan-nano-128k-Q3_K_L.gguf) | Q3_K_L | 2.24GB | false | Lower quality but usable, good for low RAM availability. | +| [Jan-nano-128k-Q3_K_M.gguf](https://huggingface.co/bartowski/Menlo_Jan-nano-128k-GGUF/blob/main/Menlo_Jan-nano-128k-Q3_K_M.gguf) | Q3_K_M | 2.08GB | false | Low quality. | +| [Jan-nano-128k-IQ3_M.gguf](https://huggingface.co/bartowski/Menlo_Jan-nano-128k-GGUF/blob/main/Menlo_Jan-nano-128k-IQ3_M.gguf) | IQ3_M | 1.96GB | false | Medium-low quality, new method with decent performance comparable to Q3_K_M. | +| [Jan-nano-128k-Q3_K_S.gguf](https://huggingface.co/bartowski/Menlo_Jan-nano-128k-GGUF/blob/main/Menlo_Jan-nano-128k-Q3_K_S.gguf) | Q3_K_S | 1.89GB | false | Low quality, not recommended. | +| [Jan-nano-128k-IQ3_XS.gguf](https://huggingface.co/bartowski/Menlo_Jan-nano-128k-GGUF/blob/main/Menlo_Jan-nano-128k-IQ3_XS.gguf) | IQ3_XS | 1.81GB | false | Lower quality, new method with decent performance, slightly better than Q3_K_S. | +| [Jan-nano-128k-Q2_K_L.gguf](https://huggingface.co/bartowski/Menlo_Jan-nano-128k-GGUF/blob/main/Menlo_Jan-nano-128k-Q2_K_L.gguf) | Q2_K_L | 1.76GB | false | Uses Q8_0 for embed and output weights. Very low quality but surprisingly usable. | +| [Jan-nano-128k-IQ3_XXS.gguf](https://huggingface.co/bartowski/Menlo_Jan-nano-128k-GGUF/blob/main/Menlo_Jan-nano-128k-IQ3_XXS.gguf) | IQ3_XXS | 1.67GB | false | Lower quality, new method with decent performance, comparable to Q3 quants. | +| [Jan-nano-128k-Q2_K.gguf](https://huggingface.co/bartowski/Menlo_Jan-nano-128k-GGUF/blob/main/Menlo_Jan-nano-128k-Q2_K.gguf) | Q2_K | 1.67GB | false | Very low quality but surprisingly usable. | +| [Jan-nano-128k-IQ2_M.gguf](https://huggingface.co/bartowski/Menlo_Jan-nano-128k-GGUF/blob/main/Menlo_Jan-nano-128k-IQ2_M.gguf) | IQ2_M | 1.51GB | 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/Menlo_Jan-nano-128k-GGUF --include "Menlo_Jan-nano-128k-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_Jan-nano-128k-GGUF --include "Menlo_Jan-nano-128k-Q8_0/*" --local-dir ./ +``` + +You can either specify a new local-dir (Menlo_Jan-nano-128k-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