bartowski/EuroLLM-9B-Instruct-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/EuroLLM-9B-Instruct-GGUF
Source: Original Platform
This commit is contained in:
ModelHub XC
2026-07-13 11:03:05 +08:00
commit a76d5f8b26
29 changed files with 346 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

61
.gitattributes vendored Normal file
View File

@@ -0,0 +1,61 @@
*.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
EuroLLM-9B-Instruct-Q8_0.gguf filter=lfs diff=lfs merge=lfs -text
EuroLLM-9B-Instruct-Q6_K_L.gguf filter=lfs diff=lfs merge=lfs -text
EuroLLM-9B-Instruct-Q6_K.gguf filter=lfs diff=lfs merge=lfs -text
EuroLLM-9B-Instruct-Q5_K_L.gguf filter=lfs diff=lfs merge=lfs -text
EuroLLM-9B-Instruct-Q5_K_M.gguf filter=lfs diff=lfs merge=lfs -text
EuroLLM-9B-Instruct-Q5_K_S.gguf filter=lfs diff=lfs merge=lfs -text
EuroLLM-9B-Instruct-Q4_K_L.gguf filter=lfs diff=lfs merge=lfs -text
EuroLLM-9B-Instruct-Q4_K_M.gguf filter=lfs diff=lfs merge=lfs -text
EuroLLM-9B-Instruct-Q4_K_S.gguf filter=lfs diff=lfs merge=lfs -text
EuroLLM-9B-Instruct-Q4_0_8_8.gguf filter=lfs diff=lfs merge=lfs -text
EuroLLM-9B-Instruct-Q4_0_4_8.gguf filter=lfs diff=lfs merge=lfs -text
EuroLLM-9B-Instruct-Q4_0_4_4.gguf filter=lfs diff=lfs merge=lfs -text
EuroLLM-9B-Instruct-Q4_0.gguf filter=lfs diff=lfs merge=lfs -text
EuroLLM-9B-Instruct-IQ4_NL.gguf filter=lfs diff=lfs merge=lfs -text
EuroLLM-9B-Instruct-IQ4_XS.gguf filter=lfs diff=lfs merge=lfs -text
EuroLLM-9B-Instruct-Q3_K_XL.gguf filter=lfs diff=lfs merge=lfs -text
EuroLLM-9B-Instruct-Q3_K_L.gguf filter=lfs diff=lfs merge=lfs -text
EuroLLM-9B-Instruct-Q3_K_M.gguf filter=lfs diff=lfs merge=lfs -text
EuroLLM-9B-Instruct-IQ3_M.gguf filter=lfs diff=lfs merge=lfs -text
EuroLLM-9B-Instruct-Q3_K_S.gguf filter=lfs diff=lfs merge=lfs -text
EuroLLM-9B-Instruct-IQ3_XS.gguf filter=lfs diff=lfs merge=lfs -text
EuroLLM-9B-Instruct-Q2_K_L.gguf filter=lfs diff=lfs merge=lfs -text
EuroLLM-9B-Instruct-Q2_K.gguf filter=lfs diff=lfs merge=lfs -text
EuroLLM-9B-Instruct-IQ2_M.gguf filter=lfs diff=lfs merge=lfs -text
EuroLLM-9B-Instruct-f16.gguf filter=lfs diff=lfs merge=lfs -text
EuroLLM-9B-Instruct.imatrix filter=lfs diff=lfs merge=lfs -text

3
EuroLLM-9B-Instruct-IQ2_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:03e5ea601ce789c38165e4e8983459d65fbd054ac18f18ce661e57cd5b03b2b5
size 3333085920

3
EuroLLM-9B-Instruct-IQ3_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:13cdfa3bdf4f5f4d54716a949a53170bfb8c0484a0878b1d219c37599e8e959a
size 4292172512

3
EuroLLM-9B-Instruct-IQ3_XS.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:416cc8665241ca28838a8dd98dfce4f2f0fb3be4781ad523016e25048ecd2c51
size 3977599712

3
EuroLLM-9B-Instruct-IQ4_NL.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:b9175884a756f3e693df6246cab05a4f480fd82d82903dd424930c2360ac1260
size 5309651680

3
EuroLLM-9B-Instruct-IQ4_XS.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:853ec21eaa2bdf58c319bcfb78fc2e11e299cda690a816b5bd71735880ecca99
size 5045541600

3
EuroLLM-9B-Instruct-Q2_K.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:df99c938ff9927e49384e57b56216b1b572df740da069d799a4ef6c184fcd92a
size 3593067232

3
EuroLLM-9B-Instruct-Q2_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:0a0c7352fbc448530b0fa7e417a96b2899b7837173a9da8556cc129c0cb41e0d
size 4105067232

3
EuroLLM-9B-Instruct-Q3_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:8b81c00366fd6e8edb1ebddfa96b5f17d6e266cd6d4984afe9b669eeb3699d3c
size 4913847008

3
EuroLLM-9B-Instruct-Q3_K_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:c4734729b75ec7eb157b9da0b5b7e946bd6ebb2b9c864b5ee1d37f399a89f8f7
size 4553136864

3
EuroLLM-9B-Instruct-Q3_K_S.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:7d7f30199702f942c67d391c94ba7a332f0a5f4edf9c243c6504e9c8ef4715e6
size 4141767392

3
EuroLLM-9B-Instruct-Q3_K_XL.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:a51adc7e8bece53b66ac88db422232857a676a09b4af603fe76e3d8e576eb04e
size 5372599008

3
EuroLLM-9B-Instruct-Q4_0.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:5b801a8e27c0222422d87f41371a554963e6129e04a2bbbb5eaac7528ed73755
size 5303360224

3
EuroLLM-9B-Instruct-Q4_0_4_4.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:c29ab685bef489507069bbabf59197ac1aa3fc435aebc2b4139b561fbc689c63
size 5287631584

3
EuroLLM-9B-Instruct-Q4_0_4_8.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:72549bf2086a4f5a0c179bdc1c9cf8e1e63ffcd31aa205ce12885933599f4b4e
size 5287631584

3
EuroLLM-9B-Instruct-Q4_0_8_8.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:5126ff453bcfa15cbe1fbbabcd469021f43459c5dc3f105a2a453abcd1dd6256
size 5287631584

3
EuroLLM-9B-Instruct-Q4_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:6d80bc7ca95bec1e19ba8f5a1d41ea629a9ea06f876893d6cfa1cfe50b20f68c
size 5971958496

3
EuroLLM-9B-Instruct-Q4_K_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:785a3b2883532381704ef74f866f822f179a931801d1ed1cf12e6deeb838806b
size 5582838496

3
EuroLLM-9B-Instruct-Q4_K_S.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:223abcb80251f117673f8622426493039153160d50a61a99288d260c80d0afae
size 5321186016

3
EuroLLM-9B-Instruct-Q5_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:1ea376f71ca1e5ed9db979d231154decfe6fb0ed8a9201ad85408a435d291456
size 6841752288

3
EuroLLM-9B-Instruct-Q5_K_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:cd04f5d39717c3d5f448dd422ecc04f7761687f54f1553830ff79aff3a33d0c0
size 6518168288

3
EuroLLM-9B-Instruct-Q5_K_S.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:c736d87c9fbe2557d184b8b645b4842e53335bde0947cf5d55a1dd46c6befa20
size 6366092000

3
EuroLLM-9B-Instruct-Q6_K.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:8b4a97dab5714ed408eb80c93dd3582ccaffbf125a77f98a668cd30a30dfe682
size 7511956192

3
EuroLLM-9B-Instruct-Q6_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:429b7ca287d9541296e7feae4325fc686d7d33e5fc94da09a98f31c1701b8d4b
size 7765908192

3
EuroLLM-9B-Instruct-Q8_0.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:5b6735847270c70eed9bb085856c3b4e5c6c74feff1e5fd270609ca8b83b85be
size 9728449248

3
EuroLLM-9B-Instruct-f16.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:5947434d0168848647e45806a3068981e3326e7da00bd4d04d2b8d71438e9ba7
size 18308422080

3
EuroLLM-9B-Instruct.imatrix Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:048130aed4575af45f5018206e793afbd4ad07e152ee61f9fe24ad230ec6b1a3
size 6202916

206
README.md Normal file
View File

@@ -0,0 +1,206 @@
---
quantized_by: bartowski
pipeline_tag: text-generation
base_model: utter-project/EuroLLM-9B-Instruct
language:
- en
- de
- es
- fr
- it
- pt
- pl
- nl
- tr
- sv
- cs
- el
- hu
- ro
- fi
- uk
- sl
- sk
- da
- lt
- lv
- et
- bg
- 'no'
- ca
- hr
- ga
- mt
- gl
- zh
- ru
- ko
- ja
- ar
- hi
license: apache-2.0
---
## Llamacpp imatrix Quantizations of EuroLLM-9B-Instruct
Using <a href="https://github.com/ggerganov/llama.cpp/">llama.cpp</a> release <a href="https://github.com/ggerganov/llama.cpp/releases/tag/b4240">b4240</a> for quantization.
Original model: https://huggingface.co/utter-project/EuroLLM-9B-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/)
## 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 |
| -------- | ---------- | --------- | ----- | ----------- |
| [EuroLLM-9B-Instruct-f16.gguf](https://huggingface.co/bartowski/EuroLLM-9B-Instruct-GGUF/blob/main/EuroLLM-9B-Instruct-f16.gguf) | f16 | 18.31GB | false | Full F16 weights. |
| [EuroLLM-9B-Instruct-Q8_0.gguf](https://huggingface.co/bartowski/EuroLLM-9B-Instruct-GGUF/blob/main/EuroLLM-9B-Instruct-Q8_0.gguf) | Q8_0 | 9.73GB | false | Extremely high quality, generally unneeded but max available quant. |
| [EuroLLM-9B-Instruct-Q6_K_L.gguf](https://huggingface.co/bartowski/EuroLLM-9B-Instruct-GGUF/blob/main/EuroLLM-9B-Instruct-Q6_K_L.gguf) | Q6_K_L | 7.77GB | false | Uses Q8_0 for embed and output weights. Very high quality, near perfect, *recommended*. |
| [EuroLLM-9B-Instruct-Q6_K.gguf](https://huggingface.co/bartowski/EuroLLM-9B-Instruct-GGUF/blob/main/EuroLLM-9B-Instruct-Q6_K.gguf) | Q6_K | 7.51GB | false | Very high quality, near perfect, *recommended*. |
| [EuroLLM-9B-Instruct-Q5_K_L.gguf](https://huggingface.co/bartowski/EuroLLM-9B-Instruct-GGUF/blob/main/EuroLLM-9B-Instruct-Q5_K_L.gguf) | Q5_K_L | 6.84GB | false | Uses Q8_0 for embed and output weights. High quality, *recommended*. |
| [EuroLLM-9B-Instruct-Q5_K_M.gguf](https://huggingface.co/bartowski/EuroLLM-9B-Instruct-GGUF/blob/main/EuroLLM-9B-Instruct-Q5_K_M.gguf) | Q5_K_M | 6.52GB | false | High quality, *recommended*. |
| [EuroLLM-9B-Instruct-Q5_K_S.gguf](https://huggingface.co/bartowski/EuroLLM-9B-Instruct-GGUF/blob/main/EuroLLM-9B-Instruct-Q5_K_S.gguf) | Q5_K_S | 6.37GB | false | High quality, *recommended*. |
| [EuroLLM-9B-Instruct-Q4_K_L.gguf](https://huggingface.co/bartowski/EuroLLM-9B-Instruct-GGUF/blob/main/EuroLLM-9B-Instruct-Q4_K_L.gguf) | Q4_K_L | 5.97GB | false | Uses Q8_0 for embed and output weights. Good quality, *recommended*. |
| [EuroLLM-9B-Instruct-Q4_K_M.gguf](https://huggingface.co/bartowski/EuroLLM-9B-Instruct-GGUF/blob/main/EuroLLM-9B-Instruct-Q4_K_M.gguf) | Q4_K_M | 5.58GB | false | Good quality, default size for most use cases, *recommended*. |
| [EuroLLM-9B-Instruct-Q3_K_XL.gguf](https://huggingface.co/bartowski/EuroLLM-9B-Instruct-GGUF/blob/main/EuroLLM-9B-Instruct-Q3_K_XL.gguf) | Q3_K_XL | 5.37GB | false | Uses Q8_0 for embed and output weights. Lower quality but usable, good for low RAM availability. |
| [EuroLLM-9B-Instruct-Q4_K_S.gguf](https://huggingface.co/bartowski/EuroLLM-9B-Instruct-GGUF/blob/main/EuroLLM-9B-Instruct-Q4_K_S.gguf) | Q4_K_S | 5.32GB | false | Slightly lower quality with more space savings, *recommended*. |
| [EuroLLM-9B-Instruct-IQ4_NL.gguf](https://huggingface.co/bartowski/EuroLLM-9B-Instruct-GGUF/blob/main/EuroLLM-9B-Instruct-IQ4_NL.gguf) | IQ4_NL | 5.31GB | false | Similar to IQ4_XS, but slightly larger. Offers online repacking for ARM CPU inference. |
| [EuroLLM-9B-Instruct-Q4_0.gguf](https://huggingface.co/bartowski/EuroLLM-9B-Instruct-GGUF/blob/main/EuroLLM-9B-Instruct-Q4_0.gguf) | Q4_0 | 5.30GB | false | Legacy format, offers online repacking for ARM CPU inference. |
| [EuroLLM-9B-Instruct-Q4_0_8_8.gguf](https://huggingface.co/bartowski/EuroLLM-9B-Instruct-GGUF/blob/main/EuroLLM-9B-Instruct-Q4_0_8_8.gguf) | Q4_0_8_8 | 5.29GB | false | Optimized for ARM and AVX inference. Requires 'sve' support for ARM (see details below). *Don't use on Mac*. |
| [EuroLLM-9B-Instruct-Q4_0_4_8.gguf](https://huggingface.co/bartowski/EuroLLM-9B-Instruct-GGUF/blob/main/EuroLLM-9B-Instruct-Q4_0_4_8.gguf) | Q4_0_4_8 | 5.29GB | false | Optimized for ARM inference. Requires 'i8mm' support (see details below). *Don't use on Mac*. |
| [EuroLLM-9B-Instruct-Q4_0_4_4.gguf](https://huggingface.co/bartowski/EuroLLM-9B-Instruct-GGUF/blob/main/EuroLLM-9B-Instruct-Q4_0_4_4.gguf) | Q4_0_4_4 | 5.29GB | false | Optimized for ARM inference. Should work well on all ARM chips, not for use with GPUs. *Don't use on Mac*. |
| [EuroLLM-9B-Instruct-IQ4_XS.gguf](https://huggingface.co/bartowski/EuroLLM-9B-Instruct-GGUF/blob/main/EuroLLM-9B-Instruct-IQ4_XS.gguf) | IQ4_XS | 5.05GB | false | Decent quality, smaller than Q4_K_S with similar performance, *recommended*. |
| [EuroLLM-9B-Instruct-Q3_K_L.gguf](https://huggingface.co/bartowski/EuroLLM-9B-Instruct-GGUF/blob/main/EuroLLM-9B-Instruct-Q3_K_L.gguf) | Q3_K_L | 4.91GB | false | Lower quality but usable, good for low RAM availability. |
| [EuroLLM-9B-Instruct-Q3_K_M.gguf](https://huggingface.co/bartowski/EuroLLM-9B-Instruct-GGUF/blob/main/EuroLLM-9B-Instruct-Q3_K_M.gguf) | Q3_K_M | 4.55GB | false | Low quality. |
| [EuroLLM-9B-Instruct-IQ3_M.gguf](https://huggingface.co/bartowski/EuroLLM-9B-Instruct-GGUF/blob/main/EuroLLM-9B-Instruct-IQ3_M.gguf) | IQ3_M | 4.29GB | false | Medium-low quality, new method with decent performance comparable to Q3_K_M. |
| [EuroLLM-9B-Instruct-Q3_K_S.gguf](https://huggingface.co/bartowski/EuroLLM-9B-Instruct-GGUF/blob/main/EuroLLM-9B-Instruct-Q3_K_S.gguf) | Q3_K_S | 4.14GB | false | Low quality, not recommended. |
| [EuroLLM-9B-Instruct-Q2_K_L.gguf](https://huggingface.co/bartowski/EuroLLM-9B-Instruct-GGUF/blob/main/EuroLLM-9B-Instruct-Q2_K_L.gguf) | Q2_K_L | 4.11GB | false | Uses Q8_0 for embed and output weights. Very low quality but surprisingly usable. |
| [EuroLLM-9B-Instruct-IQ3_XS.gguf](https://huggingface.co/bartowski/EuroLLM-9B-Instruct-GGUF/blob/main/EuroLLM-9B-Instruct-IQ3_XS.gguf) | IQ3_XS | 3.98GB | false | Lower quality, new method with decent performance, slightly better than Q3_K_S. |
| [EuroLLM-9B-Instruct-Q2_K.gguf](https://huggingface.co/bartowski/EuroLLM-9B-Instruct-GGUF/blob/main/EuroLLM-9B-Instruct-Q2_K.gguf) | Q2_K | 3.59GB | false | Very low quality but surprisingly usable. |
| [EuroLLM-9B-Instruct-IQ2_M.gguf](https://huggingface.co/bartowski/EuroLLM-9B-Instruct-GGUF/blob/main/EuroLLM-9B-Instruct-IQ2_M.gguf) | IQ2_M | 3.33GB | 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/EuroLLM-9B-Instruct-GGUF --include "EuroLLM-9B-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/EuroLLM-9B-Instruct-GGUF --include "EuroLLM-9B-Instruct-Q8_0/*" --local-dir ./
```
You can either specify a new local-dir (EuroLLM-9B-Instruct-Q8_0) or download them all in place (./)
</details>
## Q4_0_X_X information
New: Thanks to efforts made to have online repacking of weights in [this PR](https://github.com/ggerganov/llama.cpp/pull/9921), you can now just use Q4_0 if your llama.cpp has been compiled for your ARM device.
Similarly, if you want to get slightly better performance, 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</summary>
These are *NOT* for Metal (Apple) or GPU (nvidia/AMD/intel) offloading, only ARM chips (and certain AVX2/AVX512 CPUs).
If you're using an ARM chip, the Q4_0_X_X quants will have a substantial speedup. Check out Q4_0_4_4 speed comparisons [on the original pull request](https://github.com/ggerganov/llama.cpp/pull/5780#pullrequestreview-21657544660)
To check which one would work best for your ARM chip, you can check [AArch64 SoC features](https://gpages.juszkiewicz.com.pl/arm-socs-table/arm-socs.html) (thanks EloyOn!).
If you're using a CPU that supports AVX2 or AVX512 (typically server CPUs and AMD's latest Zen5 CPUs) and are not offloading to a GPU, the Q4_0_8_8 may offer a nice speed as well:
<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.
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}