bartowski/Mawdistical-S1_Infracelestial-7B-GGUF
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity

初始化项目，由ModelHub XC社区提供模型

Browse Source 
Model: bartowski/Mawdistical-S1_Infracelestial-7B-GGUF
Source: Original Platform
This commit is contained in:
ModelHub XC
2026-04-24 16:00:00 +08:00
commit 5844b4cbf7
27 changed files with 328 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

60
.gitattributes vendored Normal file
View File

@@ -0,0 +1,60 @@
*.7z filter=lfs diff=lfs merge=lfs -text
*.arrow filter=lfs diff=lfs merge=lfs -text
*.bin filter=lfs diff=lfs merge=lfs -text
*.bz2 filter=lfs diff=lfs merge=lfs -text
*.ckpt filter=lfs diff=lfs merge=lfs -text
*.ftz filter=lfs diff=lfs merge=lfs -text
*.gz filter=lfs diff=lfs merge=lfs -text
*.h5 filter=lfs diff=lfs merge=lfs -text
*.joblib filter=lfs diff=lfs merge=lfs -text
*.lfs.* filter=lfs diff=lfs merge=lfs -text
*.mlmodel filter=lfs diff=lfs merge=lfs -text
*.model filter=lfs diff=lfs merge=lfs -text
*.msgpack filter=lfs diff=lfs merge=lfs -text
*.npy filter=lfs diff=lfs merge=lfs -text
*.npz filter=lfs diff=lfs merge=lfs -text
*.onnx filter=lfs diff=lfs merge=lfs -text
*.ot filter=lfs diff=lfs merge=lfs -text
*.parquet filter=lfs diff=lfs merge=lfs -text
*.pb filter=lfs diff=lfs merge=lfs -text
*.pickle filter=lfs diff=lfs merge=lfs -text
*.pkl filter=lfs diff=lfs merge=lfs -text
*.pt filter=lfs diff=lfs merge=lfs -text
*.pth filter=lfs diff=lfs merge=lfs -text
*.rar filter=lfs diff=lfs merge=lfs -text
*.safetensors filter=lfs diff=lfs merge=lfs -text
saved_model/**/* filter=lfs diff=lfs merge=lfs -text
*.tar.* filter=lfs diff=lfs merge=lfs -text
*.tar filter=lfs diff=lfs merge=lfs -text
*.tflite filter=lfs diff=lfs merge=lfs -text
*.tgz filter=lfs diff=lfs merge=lfs -text
*.wasm filter=lfs diff=lfs merge=lfs -text
*.xz filter=lfs diff=lfs merge=lfs -text
*.zip filter=lfs diff=lfs merge=lfs -text
*.zst filter=lfs diff=lfs merge=lfs -text
*tfevents* filter=lfs diff=lfs merge=lfs -text
Mawdistical-S1_Infracelestial-7B-Q8_0.gguf filter=lfs diff=lfs merge=lfs -text
Mawdistical-S1_Infracelestial-7B-Q6_K_L.gguf filter=lfs diff=lfs merge=lfs -text
Mawdistical-S1_Infracelestial-7B-Q6_K.gguf filter=lfs diff=lfs merge=lfs -text
Mawdistical-S1_Infracelestial-7B-Q5_K_L.gguf filter=lfs diff=lfs merge=lfs -text
Mawdistical-S1_Infracelestial-7B-Q5_K_M.gguf filter=lfs diff=lfs merge=lfs -text
Mawdistical-S1_Infracelestial-7B-Q5_K_S.gguf filter=lfs diff=lfs merge=lfs -text
Mawdistical-S1_Infracelestial-7B-Q4_K_L.gguf filter=lfs diff=lfs merge=lfs -text
Mawdistical-S1_Infracelestial-7B-Q4_K_M.gguf filter=lfs diff=lfs merge=lfs -text
Mawdistical-S1_Infracelestial-7B-Q4_K_S.gguf filter=lfs diff=lfs merge=lfs -text
Mawdistical-S1_Infracelestial-7B-Q4_1.gguf filter=lfs diff=lfs merge=lfs -text
Mawdistical-S1_Infracelestial-7B-Q4_0.gguf filter=lfs diff=lfs merge=lfs -text
Mawdistical-S1_Infracelestial-7B-IQ4_NL.gguf filter=lfs diff=lfs merge=lfs -text
Mawdistical-S1_Infracelestial-7B-IQ4_XS.gguf filter=lfs diff=lfs merge=lfs -text
Mawdistical-S1_Infracelestial-7B-Q3_K_XL.gguf filter=lfs diff=lfs merge=lfs -text
Mawdistical-S1_Infracelestial-7B-Q3_K_L.gguf filter=lfs diff=lfs merge=lfs -text
Mawdistical-S1_Infracelestial-7B-Q3_K_M.gguf filter=lfs diff=lfs merge=lfs -text
Mawdistical-S1_Infracelestial-7B-IQ3_M.gguf filter=lfs diff=lfs merge=lfs -text
Mawdistical-S1_Infracelestial-7B-Q3_K_S.gguf filter=lfs diff=lfs merge=lfs -text
Mawdistical-S1_Infracelestial-7B-IQ3_XS.gguf filter=lfs diff=lfs merge=lfs -text
Mawdistical-S1_Infracelestial-7B-IQ3_XXS.gguf filter=lfs diff=lfs merge=lfs -text
Mawdistical-S1_Infracelestial-7B-Q2_K_L.gguf filter=lfs diff=lfs merge=lfs -text
Mawdistical-S1_Infracelestial-7B-Q2_K.gguf filter=lfs diff=lfs merge=lfs -text
Mawdistical-S1_Infracelestial-7B-IQ2_M.gguf filter=lfs diff=lfs merge=lfs -text
Mawdistical-S1_Infracelestial-7B-bf16.gguf filter=lfs diff=lfs merge=lfs -text
Mawdistical-S1_Infracelestial-7B-imatrix.gguf filter=lfs diff=lfs merge=lfs -text

3
Mawdistical-S1_Infracelestial-7B-IQ2_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:6bf970ca61fa03c3e9603ae541b1614c0b17854e27097fc8d61c2dd534c520e5
size 3168859456

3
Mawdistical-S1_Infracelestial-7B-IQ3_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:625895f4de9ad79669775c9f1df04db5880702ac60dddae81f210565a9f9c6b2
size 3786900800

3
Mawdistical-S1_Infracelestial-7B-IQ3_XS.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:9767c76913f17af745bd21816a5485eca2be0d3207c191b47ba058aa9e0a5cb5
size 3549726016

3
Mawdistical-S1_Infracelestial-7B-IQ3_XXS.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:1156367dcbd36b02586e4e0caefa022435cbd3c6f1625b6b2d63a29e5a866cb1
size 3331847488

3
Mawdistical-S1_Infracelestial-7B-IQ4_NL.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:c495d255535c74ef2050a76048ca7e2bb869775133b0095af41f18385bdc95fd
size 4572419392

3
Mawdistical-S1_Infracelestial-7B-IQ4_XS.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:d27c35b30c3d8e093345c9efddffd6dc1fd3b98877e4a76f63aad53472905789
size 4367799616

3
Mawdistical-S1_Infracelestial-7B-Q2_K.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:6a38b31cec0f1c26c15cc53728b0b693dd1b6b298efdf1da499d724b97813a22
size 3225040192

3
Mawdistical-S1_Infracelestial-7B-Q2_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:ee19ce124ff29335b7a1ffdd7212456e29c89ff0204aa62e7f1153dba1343b7a
size 3831760192

3
Mawdistical-S1_Infracelestial-7B-Q3_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:a3cdca88b2a86c22e4843c72e12a5584040f2b2cfb5a1262ddfc98c5a38e7694
size 4181427520

3
Mawdistical-S1_Infracelestial-7B-Q3_K_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:09236beac44d0134bd41763e18144251112604f34b29c7a46f93c7de57c41353
size 3980363072

3
Mawdistical-S1_Infracelestial-7B-Q3_K_S.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:bef476fe344d270ae38258d5e80a5a13fa49e062326c954ee0a0f53dec7fcaa6
size 3643802944

3
Mawdistical-S1_Infracelestial-7B-Q3_K_XL.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:ad0612f1025a8e00db25f94c2f90ba8f7cafbe2c7b27a2abe7dc8168b149f474
size 4725048640

3
Mawdistical-S1_Infracelestial-7B-Q4_0.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:c02eeefb40a31439128483a4a49c431efcb29b1c5844807821c832145dee46f1
size 4544763200

3
Mawdistical-S1_Infracelestial-7B-Q4_1.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:c171df2d80ad9d13a0393dadedaab6ee3a1c97a6bc07e2768201960155dd0e21
size 4952167744

3
Mawdistical-S1_Infracelestial-7B-Q4_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:5fe60421b9ffe597541ac3adbc4140caba818def51357a441d1e700a34c3bf2d
size 5317674304

3
Mawdistical-S1_Infracelestial-7B-Q4_K_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:2cc112427d3d799e757339399b441422bb0fe5d7d9a4ae801f921143d6eebb6c
size 4856567104

3
Mawdistical-S1_Infracelestial-7B-Q4_K_S.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:0ff0bb4446f625ff469c268704293fe5c967bc556b99335e866e7189c51afba3
size 4565472576

3
Mawdistical-S1_Infracelestial-7B-Q5_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:84e71a2d86bfdfc2f0570319e684d3c389618fd6ace4e6d0be52ff120f3d3995
size 5972378944

3
Mawdistical-S1_Infracelestial-7B-Q5_K_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:772fec3edfe44e6f1c749cbec3f7fdcafe3a94b436d6d767f0da272a518c1564
size 5588931904

3
Mawdistical-S1_Infracelestial-7B-Q5_K_S.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:f68d9111dff7c37a21ca331579f1cccf5428837a876a2228db1620bd1ebfb4cf
size 5370844480

3
Mawdistical-S1_Infracelestial-7B-Q6_K.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:14333b283e1cd12232d6243eeead17193fde2650852625537dae70925aacc7c2
size 6534800704

3
Mawdistical-S1_Infracelestial-7B-Q6_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:0741c02c9ad3bcbc955ed4a14eeaeb17898f20ae7b404e72403605db6457f20b
size 6835733824

3
Mawdistical-S1_Infracelestial-7B-Q8_0.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:7633fd88076128315aeda2bf816a58da025f9689e9691cb97b5ec6f3a8287f74
size 8106509632

3
Mawdistical-S1_Infracelestial-7B-bf16.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:19f495cd207da82bba2042b33bb4bf3681378cad14da9daf0301add2f35be767
size 15252227072

3
Mawdistical-S1_Infracelestial-7B-imatrix.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:9688016f634fa7b4a4c30d23ba90673cb2fb7d9f8621ae277880cd1b2f439ef4
size 5162880

193
README.md Normal file
View File

@@ -0,0 +1,193 @@
---
quantized_by: bartowski
pipeline_tag: text-generation
tags:
- nsfw
- explicit
- roleplay
- mixed-AI
- furry
- anthro
- chat
- manipulation
- sfw
license: other
language:
- en
base_model_relation: quantized
inference: false
thumbnail: https://cdn-uploads.huggingface.co/production/uploads/67c10cfba43d7939d60160ff/jk4OFD0hJma2GT9vTqlRR.jpeg
base_model: Mawdistical-S1/Infracelestial-7B
---
## Llamacpp imatrix Quantizations of Infracelestial-7B by Mawdistical-S1
Using <a href="https://github.com/ggml-org/llama.cpp/">llama.cpp</a> release <a href="https://github.com/ggml-org/llama.cpp/releases/tag/b8779">b8779</a> for quantization.
Original model: https://huggingface.co/Mawdistical-S1/Infracelestial-7B
All quants made using imatrix option with dataset from [here](https://gist.github.com/bartowski1182/82ae9b520227f57d79ba04add13d0d0d)
Run them in your choice of tools:
- [llama.cpp](https://github.com/ggml-org/llama.cpp)
- [ramalama](https://github.com/containers/ramalama)
- [LM Studio](https://lmstudio.ai/)
- [koboldcpp](https://github.com/LostRuins/koboldcpp)
- [Jan AI](https://www.jan.ai/)
- [Text Generation Web UI](https://github.com/oobabooga/text-generation-webui)
- [LoLLMs](https://github.com/ParisNeo/lollms)
Note: if it's a newly supported model, you may need to wait for an update from the developers.
## Prompt format
```
<|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 |
| -------- | ---------- | --------- | ----- | ----------- |
| [Infracelestial-7B-bf16.gguf](https://huggingface.co/bartowski/Mawdistical-S1_Infracelestial-7B-GGUF/blob/main/Mawdistical-S1_Infracelestial-7B-bf16.gguf) | bf16 | 15.25GB | false | Full BF16 weights. |
| [Infracelestial-7B-Q8_0.gguf](https://huggingface.co/bartowski/Mawdistical-S1_Infracelestial-7B-GGUF/blob/main/Mawdistical-S1_Infracelestial-7B-Q8_0.gguf) | Q8_0 | 8.11GB | false | Extremely high quality, generally unneeded but max available quant. |
| [Infracelestial-7B-Q6_K_L.gguf](https://huggingface.co/bartowski/Mawdistical-S1_Infracelestial-7B-GGUF/blob/main/Mawdistical-S1_Infracelestial-7B-Q6_K_L.gguf) | Q6_K_L | 6.84GB | false | Uses Q8_0 for embed and output weights. Very high quality, near perfect, *recommended*. |
| [Infracelestial-7B-Q6_K.gguf](https://huggingface.co/bartowski/Mawdistical-S1_Infracelestial-7B-GGUF/blob/main/Mawdistical-S1_Infracelestial-7B-Q6_K.gguf) | Q6_K | 6.53GB | false | Very high quality, near perfect, *recommended*. |
| [Infracelestial-7B-Q5_K_L.gguf](https://huggingface.co/bartowski/Mawdistical-S1_Infracelestial-7B-GGUF/blob/main/Mawdistical-S1_Infracelestial-7B-Q5_K_L.gguf) | Q5_K_L | 5.97GB | false | Uses Q8_0 for embed and output weights. High quality, *recommended*. |
| [Infracelestial-7B-Q5_K_M.gguf](https://huggingface.co/bartowski/Mawdistical-S1_Infracelestial-7B-GGUF/blob/main/Mawdistical-S1_Infracelestial-7B-Q5_K_M.gguf) | Q5_K_M | 5.59GB | false | High quality, *recommended*. |
| [Infracelestial-7B-Q5_K_S.gguf](https://huggingface.co/bartowski/Mawdistical-S1_Infracelestial-7B-GGUF/blob/main/Mawdistical-S1_Infracelestial-7B-Q5_K_S.gguf) | Q5_K_S | 5.37GB | false | High quality, *recommended*. |
| [Infracelestial-7B-Q4_K_L.gguf](https://huggingface.co/bartowski/Mawdistical-S1_Infracelestial-7B-GGUF/blob/main/Mawdistical-S1_Infracelestial-7B-Q4_K_L.gguf) | Q4_K_L | 5.32GB | false | Uses Q8_0 for embed and output weights. Good quality, *recommended*. |
| [Infracelestial-7B-Q4_1.gguf](https://huggingface.co/bartowski/Mawdistical-S1_Infracelestial-7B-GGUF/blob/main/Mawdistical-S1_Infracelestial-7B-Q4_1.gguf) | Q4_1 | 4.95GB | false | Legacy format, similar performance to Q4_K_S but with improved tokens/watt on Apple silicon. |
| [Infracelestial-7B-Q4_K_M.gguf](https://huggingface.co/bartowski/Mawdistical-S1_Infracelestial-7B-GGUF/blob/main/Mawdistical-S1_Infracelestial-7B-Q4_K_M.gguf) | Q4_K_M | 4.86GB | false | Good quality, default size for most use cases, *recommended*. |
| [Infracelestial-7B-Q3_K_XL.gguf](https://huggingface.co/bartowski/Mawdistical-S1_Infracelestial-7B-GGUF/blob/main/Mawdistical-S1_Infracelestial-7B-Q3_K_XL.gguf) | Q3_K_XL | 4.73GB | false | Uses Q8_0 for embed and output weights. Lower quality but usable, good for low RAM availability. |
| [Infracelestial-7B-Q4_K_S.gguf](https://huggingface.co/bartowski/Mawdistical-S1_Infracelestial-7B-GGUF/blob/main/Mawdistical-S1_Infracelestial-7B-Q4_K_S.gguf) | Q4_K_S | 4.57GB | false | Slightly lower quality with more space savings, *recommended*. |
| [Infracelestial-7B-IQ4_NL.gguf](https://huggingface.co/bartowski/Mawdistical-S1_Infracelestial-7B-GGUF/blob/main/Mawdistical-S1_Infracelestial-7B-IQ4_NL.gguf) | IQ4_NL | 4.57GB | false | Similar to IQ4_XS, but slightly larger. Offers online repacking for ARM CPU inference. |
| [Infracelestial-7B-Q4_0.gguf](https://huggingface.co/bartowski/Mawdistical-S1_Infracelestial-7B-GGUF/blob/main/Mawdistical-S1_Infracelestial-7B-Q4_0.gguf) | Q4_0 | 4.54GB | false | Legacy format, offers online repacking for ARM and AVX CPU inference. |
| [Infracelestial-7B-IQ4_XS.gguf](https://huggingface.co/bartowski/Mawdistical-S1_Infracelestial-7B-GGUF/blob/main/Mawdistical-S1_Infracelestial-7B-IQ4_XS.gguf) | IQ4_XS | 4.37GB | false | Decent quality, smaller than Q4_K_S with similar performance, *recommended*. |
| [Infracelestial-7B-Q3_K_L.gguf](https://huggingface.co/bartowski/Mawdistical-S1_Infracelestial-7B-GGUF/blob/main/Mawdistical-S1_Infracelestial-7B-Q3_K_L.gguf) | Q3_K_L | 4.18GB | false | Lower quality but usable, good for low RAM availability. |
| [Infracelestial-7B-Q3_K_M.gguf](https://huggingface.co/bartowski/Mawdistical-S1_Infracelestial-7B-GGUF/blob/main/Mawdistical-S1_Infracelestial-7B-Q3_K_M.gguf) | Q3_K_M | 3.98GB | false | Low quality. |
| [Infracelestial-7B-Q2_K_L.gguf](https://huggingface.co/bartowski/Mawdistical-S1_Infracelestial-7B-GGUF/blob/main/Mawdistical-S1_Infracelestial-7B-Q2_K_L.gguf) | Q2_K_L | 3.83GB | false | Uses Q8_0 for embed and output weights. Very low quality but surprisingly usable. |
| [Infracelestial-7B-IQ3_M.gguf](https://huggingface.co/bartowski/Mawdistical-S1_Infracelestial-7B-GGUF/blob/main/Mawdistical-S1_Infracelestial-7B-IQ3_M.gguf) | IQ3_M | 3.79GB | false | Medium-low quality, new method with decent performance comparable to Q3_K_M. |
| [Infracelestial-7B-Q3_K_S.gguf](https://huggingface.co/bartowski/Mawdistical-S1_Infracelestial-7B-GGUF/blob/main/Mawdistical-S1_Infracelestial-7B-Q3_K_S.gguf) | Q3_K_S | 3.64GB | false | Low quality, not recommended. |
| [Infracelestial-7B-IQ3_XS.gguf](https://huggingface.co/bartowski/Mawdistical-S1_Infracelestial-7B-GGUF/blob/main/Mawdistical-S1_Infracelestial-7B-IQ3_XS.gguf) | IQ3_XS | 3.55GB | false | Lower quality, new method with decent performance, slightly better than Q3_K_S. |
| [Infracelestial-7B-IQ3_XXS.gguf](https://huggingface.co/bartowski/Mawdistical-S1_Infracelestial-7B-GGUF/blob/main/Mawdistical-S1_Infracelestial-7B-IQ3_XXS.gguf) | IQ3_XXS | 3.33GB | false | Lower quality, new method with decent performance, comparable to Q3 quants. |
| [Infracelestial-7B-Q2_K.gguf](https://huggingface.co/bartowski/Mawdistical-S1_Infracelestial-7B-GGUF/blob/main/Mawdistical-S1_Infracelestial-7B-Q2_K.gguf) | Q2_K | 3.23GB | false | Very low quality but surprisingly usable. |
| [Infracelestial-7B-IQ2_M.gguf](https://huggingface.co/bartowski/Mawdistical-S1_Infracelestial-7B-GGUF/blob/main/Mawdistical-S1_Infracelestial-7B-IQ2_M.gguf) | IQ2_M | 3.17GB | 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
<details>
<summary>Click to view download instructions</summary>
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/Mawdistical-S1_Infracelestial-7B-GGUF --include "Mawdistical-S1_Infracelestial-7B-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/Mawdistical-S1_Infracelestial-7B-GGUF --include "Mawdistical-S1_Infracelestial-7B-Q8_0/*" --local-dir ./
```
You can either specify a new local-dir (Mawdistical-S1_Infracelestial-7B-Q8_0) or download them all in place (./)
</details>
## 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.
<details>
<summary>Click to view Q4_0_X_X information (deprecated</summary>
I'm keeping this section to show the potential theoretical uplift in performance from using the Q4_0 with online repacking.
<details>
<summary>Click to view benchmarks on an AVX2 system (EPYC7702)</summary>
| 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
</details>
</details>
## Which file should I choose?
<details>
<summary>Click here for details</summary>
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.
</details>
## 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