bartowski/PygmalionAI_Eleusis-12B-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/PygmalionAI_Eleusis-12B-GGUF
Source: Original Platform
This commit is contained in:
ModelHub XC
2026-08-25 19:50:13 +08:00
commit b9bbd91e56
28 changed files with 308 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
PygmalionAI_Eleusis-12B-f16.gguf filter=lfs diff=lfs merge=lfs -text
PygmalionAI_Eleusis-12B-Q8_0.gguf filter=lfs diff=lfs merge=lfs -text
PygmalionAI_Eleusis-12B-Q6_K_L.gguf filter=lfs diff=lfs merge=lfs -text
PygmalionAI_Eleusis-12B-Q6_K.gguf filter=lfs diff=lfs merge=lfs -text
PygmalionAI_Eleusis-12B-Q5_K_L.gguf filter=lfs diff=lfs merge=lfs -text
PygmalionAI_Eleusis-12B-Q5_K_M.gguf filter=lfs diff=lfs merge=lfs -text
PygmalionAI_Eleusis-12B-Q5_K_S.gguf filter=lfs diff=lfs merge=lfs -text
PygmalionAI_Eleusis-12B-Q4_K_L.gguf filter=lfs diff=lfs merge=lfs -text
PygmalionAI_Eleusis-12B-Q4_K_M.gguf filter=lfs diff=lfs merge=lfs -text
PygmalionAI_Eleusis-12B-Q4_K_S.gguf filter=lfs diff=lfs merge=lfs -text
PygmalionAI_Eleusis-12B-Q4_1.gguf filter=lfs diff=lfs merge=lfs -text
PygmalionAI_Eleusis-12B-Q4_0.gguf filter=lfs diff=lfs merge=lfs -text
PygmalionAI_Eleusis-12B-IQ4_NL.gguf filter=lfs diff=lfs merge=lfs -text
PygmalionAI_Eleusis-12B-IQ4_XS.gguf filter=lfs diff=lfs merge=lfs -text
PygmalionAI_Eleusis-12B-Q3_K_XL.gguf filter=lfs diff=lfs merge=lfs -text
PygmalionAI_Eleusis-12B-Q3_K_L.gguf filter=lfs diff=lfs merge=lfs -text
PygmalionAI_Eleusis-12B-Q3_K_M.gguf filter=lfs diff=lfs merge=lfs -text
PygmalionAI_Eleusis-12B-IQ3_M.gguf filter=lfs diff=lfs merge=lfs -text
PygmalionAI_Eleusis-12B-Q3_K_S.gguf filter=lfs diff=lfs merge=lfs -text
PygmalionAI_Eleusis-12B-IQ3_XS.gguf filter=lfs diff=lfs merge=lfs -text
PygmalionAI_Eleusis-12B-IQ3_XXS.gguf filter=lfs diff=lfs merge=lfs -text
PygmalionAI_Eleusis-12B-Q2_K_L.gguf filter=lfs diff=lfs merge=lfs -text
PygmalionAI_Eleusis-12B-Q2_K.gguf filter=lfs diff=lfs merge=lfs -text
PygmalionAI_Eleusis-12B-IQ2_M.gguf filter=lfs diff=lfs merge=lfs -text
PygmalionAI_Eleusis-12B.imatrix filter=lfs diff=lfs merge=lfs -text

3
PygmalionAI_Eleusis-12B-IQ2_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:7e04974e247de5d493dce9c37c4f0125ff473fa105960566ea3110034b8b6047
size 4435023680

3
PygmalionAI_Eleusis-12B-IQ3_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:230eedc79cc04831837f009bf9cb659e6e3bb3c42b5d90569f77c53f5952f44e
size 5722232640

3
PygmalionAI_Eleusis-12B-IQ3_XS.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:903391ca84c1197178f99903671475bfd2c86cd06080746ef0a0a2c45e3aaf78
size 5306488640

3
PygmalionAI_Eleusis-12B-IQ3_XXS.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:3984225a62b12cfb8339081c2015f32b4e060edb543ad4794ec50f78574b5853
size 4945385280

3
PygmalionAI_Eleusis-12B-IQ4_NL.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:4f9735391c75aac6983469e21e88abb392c11aeebaad14b943934b34fb22bcc8
size 7097915200

3
PygmalionAI_Eleusis-12B-IQ4_XS.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:9be258929744d96922737852a9ad4028c3e231e673c550efbe6aa8261bb98fc2
size 6742710080

3
PygmalionAI_Eleusis-12B-Q2_K.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:141076684bb2584bc4d46d8a4dd8e75bd1b750c9d8bd6bc82f29a2752a14b1ad
size 4791048000

3
PygmalionAI_Eleusis-12B-Q2_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:45372c1e6ca7f5c952bf2f0cea551c569a997a3a1ff97d912c32fd488ee56a34
size 5446408000

3
PygmalionAI_Eleusis-12B-Q3_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:7b20240b655e8d78f0430609d36c2dbcbe460c9b908b8b2144be576a35b92f69
size 6561503040

3
PygmalionAI_Eleusis-12B-Q3_K_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:27c632016db447d1ee85a5749ed2498fdbffcd9cac408272eb8399cd81acfb3a
size 6083090240

3
PygmalionAI_Eleusis-12B-Q3_K_S.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:257caca22e870b37565c70b52a0bbcbd3c658148254817260c02494a0c5e543e
size 5534226240

3
PygmalionAI_Eleusis-12B-Q3_K_XL.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:031a1560709cd72ab4d9d9d92bf132b38827871c0409100acfd800c2c1bbee80
size 7148705600

3
PygmalionAI_Eleusis-12B-Q4_0.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:61946a6087e0aa774fdc7c23ebeb2657338908da4c286564ac1485756a1cf960
size 7094638400

3
PygmalionAI_Eleusis-12B-Q4_1.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:b9457b854b8aadfc24c069420f205ab16aade0ed519525496ac7c7de3c8dcc60
size 7795218240

3
PygmalionAI_Eleusis-12B-Q4_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:e96a3037032a2576c36ae83d1cae0b4d9799033f6dcfd408dd7b8afb7ab04700
size 7975278400

3
PygmalionAI_Eleusis-12B-Q4_K_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:899091671ae483fc7c132512221ee6600984c936cd8c261becee696d00080701
size 7477204800

3
PygmalionAI_Eleusis-12B-Q4_K_S.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:2b5d1342909f63f65dcb490b515a35001ec2255f115cd62fb044b020c7c93eac
size 7120197440

3
PygmalionAI_Eleusis-12B-Q5_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:365d4358c3a5514ec3e33a1a7afa71f66448c21dc71e7e23889684ea2ffe625b
size 9141819200

3
PygmalionAI_Eleusis-12B-Q5_K_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:3b5b1dbe8f1f1cdccdbcf52c0678b41929691772a0bb66ab6d249729e216a1b4
size 8727631680

3
PygmalionAI_Eleusis-12B-Q5_K_S.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:c8291be0520709b341ca09727ec2d2bd18139ff3d2538a5009aac84ad8f60889
size 8518735680

3
PygmalionAI_Eleusis-12B-Q6_K.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:5542d1e6f58a6ddb2f576556443563a7578e969f869751992842011ffe1b4502
size 10056210240

3
PygmalionAI_Eleusis-12B-Q6_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:7ae9b33f4cdb6c135cfa4d9b1fa9afeaea2e1027e9486d2c705db82667e5a1b0
size 10381268800

3
PygmalionAI_Eleusis-12B-Q8_0.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:f346a18cde1803b2cd561ee9c511720732ad90d5030cb536ef5e548ae49c3357
size 13022369600

3
PygmalionAI_Eleusis-12B-f16.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:98f3c800527282d86322560e4c4be92f4d46a6c90ce0d5c5d4456a5c6053323b
size 24504276800

3
PygmalionAI_Eleusis-12B.imatrix Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:4bc0a7ccd7f6cbb369a11f11aa352324231ed67316047f3eaadbf2e8166c4470
size 7054418

172
README.md Normal file
View File

@@ -0,0 +1,172 @@
---
quantized_by: bartowski
pipeline_tag: text-generation
license: apache-2.0
base_model: PygmalionAI/Eleusis-12B
---
## Llamacpp imatrix Quantizations of Eleusis-12B by PygmalionAI
Using <a href="https://github.com/ggerganov/llama.cpp/">llama.cpp</a> release <a href="https://github.com/ggerganov/llama.cpp/releases/tag/b4688">b4688</a> for quantization.
Original model: https://huggingface.co/PygmalionAI/Eleusis-12B
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 |
| -------- | ---------- | --------- | ----- | ----------- |
| [Eleusis-12B-f16.gguf](https://huggingface.co/bartowski/PygmalionAI_Eleusis-12B-GGUF/blob/main/PygmalionAI_Eleusis-12B-f16.gguf) | f16 | 24.50GB | false | Full F16 weights. |
| [Eleusis-12B-Q8_0.gguf](https://huggingface.co/bartowski/PygmalionAI_Eleusis-12B-GGUF/blob/main/PygmalionAI_Eleusis-12B-Q8_0.gguf) | Q8_0 | 13.02GB | false | Extremely high quality, generally unneeded but max available quant. |
| [Eleusis-12B-Q6_K_L.gguf](https://huggingface.co/bartowski/PygmalionAI_Eleusis-12B-GGUF/blob/main/PygmalionAI_Eleusis-12B-Q6_K_L.gguf) | Q6_K_L | 10.38GB | false | Uses Q8_0 for embed and output weights. Very high quality, near perfect, *recommended*. |
| [Eleusis-12B-Q6_K.gguf](https://huggingface.co/bartowski/PygmalionAI_Eleusis-12B-GGUF/blob/main/PygmalionAI_Eleusis-12B-Q6_K.gguf) | Q6_K | 10.06GB | false | Very high quality, near perfect, *recommended*. |
| [Eleusis-12B-Q5_K_L.gguf](https://huggingface.co/bartowski/PygmalionAI_Eleusis-12B-GGUF/blob/main/PygmalionAI_Eleusis-12B-Q5_K_L.gguf) | Q5_K_L | 9.14GB | false | Uses Q8_0 for embed and output weights. High quality, *recommended*. |
| [Eleusis-12B-Q5_K_M.gguf](https://huggingface.co/bartowski/PygmalionAI_Eleusis-12B-GGUF/blob/main/PygmalionAI_Eleusis-12B-Q5_K_M.gguf) | Q5_K_M | 8.73GB | false | High quality, *recommended*. |
| [Eleusis-12B-Q5_K_S.gguf](https://huggingface.co/bartowski/PygmalionAI_Eleusis-12B-GGUF/blob/main/PygmalionAI_Eleusis-12B-Q5_K_S.gguf) | Q5_K_S | 8.52GB | false | High quality, *recommended*. |
| [Eleusis-12B-Q4_K_L.gguf](https://huggingface.co/bartowski/PygmalionAI_Eleusis-12B-GGUF/blob/main/PygmalionAI_Eleusis-12B-Q4_K_L.gguf) | Q4_K_L | 7.98GB | false | Uses Q8_0 for embed and output weights. Good quality, *recommended*. |
| [Eleusis-12B-Q4_1.gguf](https://huggingface.co/bartowski/PygmalionAI_Eleusis-12B-GGUF/blob/main/PygmalionAI_Eleusis-12B-Q4_1.gguf) | Q4_1 | 7.80GB | false | Legacy format, similar performance to Q4_K_S but with improved tokens/watt on Apple silicon. |
| [Eleusis-12B-Q4_K_M.gguf](https://huggingface.co/bartowski/PygmalionAI_Eleusis-12B-GGUF/blob/main/PygmalionAI_Eleusis-12B-Q4_K_M.gguf) | Q4_K_M | 7.48GB | false | Good quality, default size for most use cases, *recommended*. |
| [Eleusis-12B-Q3_K_XL.gguf](https://huggingface.co/bartowski/PygmalionAI_Eleusis-12B-GGUF/blob/main/PygmalionAI_Eleusis-12B-Q3_K_XL.gguf) | Q3_K_XL | 7.15GB | false | Uses Q8_0 for embed and output weights. Lower quality but usable, good for low RAM availability. |
| [Eleusis-12B-Q4_K_S.gguf](https://huggingface.co/bartowski/PygmalionAI_Eleusis-12B-GGUF/blob/main/PygmalionAI_Eleusis-12B-Q4_K_S.gguf) | Q4_K_S | 7.12GB | false | Slightly lower quality with more space savings, *recommended*. |
| [Eleusis-12B-IQ4_NL.gguf](https://huggingface.co/bartowski/PygmalionAI_Eleusis-12B-GGUF/blob/main/PygmalionAI_Eleusis-12B-IQ4_NL.gguf) | IQ4_NL | 7.10GB | false | Similar to IQ4_XS, but slightly larger. Offers online repacking for ARM CPU inference. |
| [Eleusis-12B-Q4_0.gguf](https://huggingface.co/bartowski/PygmalionAI_Eleusis-12B-GGUF/blob/main/PygmalionAI_Eleusis-12B-Q4_0.gguf) | Q4_0 | 7.09GB | false | Legacy format, offers online repacking for ARM and AVX CPU inference. |
| [Eleusis-12B-IQ4_XS.gguf](https://huggingface.co/bartowski/PygmalionAI_Eleusis-12B-GGUF/blob/main/PygmalionAI_Eleusis-12B-IQ4_XS.gguf) | IQ4_XS | 6.74GB | false | Decent quality, smaller than Q4_K_S with similar performance, *recommended*. |
| [Eleusis-12B-Q3_K_L.gguf](https://huggingface.co/bartowski/PygmalionAI_Eleusis-12B-GGUF/blob/main/PygmalionAI_Eleusis-12B-Q3_K_L.gguf) | Q3_K_L | 6.56GB | false | Lower quality but usable, good for low RAM availability. |
| [Eleusis-12B-Q3_K_M.gguf](https://huggingface.co/bartowski/PygmalionAI_Eleusis-12B-GGUF/blob/main/PygmalionAI_Eleusis-12B-Q3_K_M.gguf) | Q3_K_M | 6.08GB | false | Low quality. |
| [Eleusis-12B-IQ3_M.gguf](https://huggingface.co/bartowski/PygmalionAI_Eleusis-12B-GGUF/blob/main/PygmalionAI_Eleusis-12B-IQ3_M.gguf) | IQ3_M | 5.72GB | false | Medium-low quality, new method with decent performance comparable to Q3_K_M. |
| [Eleusis-12B-Q3_K_S.gguf](https://huggingface.co/bartowski/PygmalionAI_Eleusis-12B-GGUF/blob/main/PygmalionAI_Eleusis-12B-Q3_K_S.gguf) | Q3_K_S | 5.53GB | false | Low quality, not recommended. |
| [Eleusis-12B-Q2_K_L.gguf](https://huggingface.co/bartowski/PygmalionAI_Eleusis-12B-GGUF/blob/main/PygmalionAI_Eleusis-12B-Q2_K_L.gguf) | Q2_K_L | 5.45GB | false | Uses Q8_0 for embed and output weights. Very low quality but surprisingly usable. |
| [Eleusis-12B-IQ3_XS.gguf](https://huggingface.co/bartowski/PygmalionAI_Eleusis-12B-GGUF/blob/main/PygmalionAI_Eleusis-12B-IQ3_XS.gguf) | IQ3_XS | 5.31GB | false | Lower quality, new method with decent performance, slightly better than Q3_K_S. |
| [Eleusis-12B-IQ3_XXS.gguf](https://huggingface.co/bartowski/PygmalionAI_Eleusis-12B-GGUF/blob/main/PygmalionAI_Eleusis-12B-IQ3_XXS.gguf) | IQ3_XXS | 4.95GB | false | Lower quality, new method with decent performance, comparable to Q3 quants. |
| [Eleusis-12B-Q2_K.gguf](https://huggingface.co/bartowski/PygmalionAI_Eleusis-12B-GGUF/blob/main/PygmalionAI_Eleusis-12B-Q2_K.gguf) | Q2_K | 4.79GB | false | Very low quality but surprisingly usable. |
| [Eleusis-12B-IQ2_M.gguf](https://huggingface.co/bartowski/PygmalionAI_Eleusis-12B-GGUF/blob/main/PygmalionAI_Eleusis-12B-IQ2_M.gguf) | IQ2_M | 4.44GB | 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/PygmalionAI_Eleusis-12B-GGUF --include "PygmalionAI_Eleusis-12B-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/PygmalionAI_Eleusis-12B-GGUF --include "PygmalionAI_Eleusis-12B-Q8_0/*" --local-dir ./
```
You can either specify a new local-dir (PygmalionAI_Eleusis-12B-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/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.
<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/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 and Apple Metal, but will be slower than their K-quant equivalent, so speed vs performance is a tradeoff you'll have to decide.
The I-quants are *not* compatible with Vulcan, which is also AMD, so if you have an AMD card double check if you're using the rocBLAS build or the Vulcan build. At the time of writing this, LM Studio has a preview with ROCm support, and other inference engines have specific builds for ROCm.
</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

1
configuration.json Normal file
View File

@@ -0,0 +1 @@
{"framework": "pytorch", "task": "text-generation", "allow_remote": true}