bartowski/EXAONE-3.5-2.4B-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/EXAONE-3.5-2.4B-Instruct-GGUF
Source: Original Platform
This commit is contained in:
ModelHub XC
2026-07-18 02:51:06 +08:00
commit 704464cb37
30 changed files with 322 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

62
.gitattributes vendored Normal file
View File

@@ -0,0 +1,62 @@
*.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
EXAONE-3.5-2.4B-Instruct-f16.gguf filter=lfs diff=lfs merge=lfs -text
EXAONE-3.5-2.4B-Instruct-Q8_0.gguf filter=lfs diff=lfs merge=lfs -text
EXAONE-3.5-2.4B-Instruct-Q6_K_L.gguf filter=lfs diff=lfs merge=lfs -text
EXAONE-3.5-2.4B-Instruct-Q6_K.gguf filter=lfs diff=lfs merge=lfs -text
EXAONE-3.5-2.4B-Instruct-Q5_K_L.gguf filter=lfs diff=lfs merge=lfs -text
EXAONE-3.5-2.4B-Instruct-Q5_K_M.gguf filter=lfs diff=lfs merge=lfs -text
EXAONE-3.5-2.4B-Instruct-Q5_K_S.gguf filter=lfs diff=lfs merge=lfs -text
EXAONE-3.5-2.4B-Instruct-Q4_K_L.gguf filter=lfs diff=lfs merge=lfs -text
EXAONE-3.5-2.4B-Instruct-Q4_K_M.gguf filter=lfs diff=lfs merge=lfs -text
EXAONE-3.5-2.4B-Instruct-Q4_K_S.gguf filter=lfs diff=lfs merge=lfs -text
EXAONE-3.5-2.4B-Instruct-Q4_0_8_8.gguf filter=lfs diff=lfs merge=lfs -text
EXAONE-3.5-2.4B-Instruct-Q4_0_4_8.gguf filter=lfs diff=lfs merge=lfs -text
EXAONE-3.5-2.4B-Instruct-Q4_0_4_4.gguf filter=lfs diff=lfs merge=lfs -text
EXAONE-3.5-2.4B-Instruct-Q4_0.gguf filter=lfs diff=lfs merge=lfs -text
EXAONE-3.5-2.4B-Instruct-IQ4_NL.gguf filter=lfs diff=lfs merge=lfs -text
EXAONE-3.5-2.4B-Instruct-IQ4_XS.gguf filter=lfs diff=lfs merge=lfs -text
EXAONE-3.5-2.4B-Instruct-Q3_K_XL.gguf filter=lfs diff=lfs merge=lfs -text
EXAONE-3.5-2.4B-Instruct-Q3_K_L.gguf filter=lfs diff=lfs merge=lfs -text
EXAONE-3.5-2.4B-Instruct-Q3_K_M.gguf filter=lfs diff=lfs merge=lfs -text
EXAONE-3.5-2.4B-Instruct-IQ3_M.gguf filter=lfs diff=lfs merge=lfs -text
EXAONE-3.5-2.4B-Instruct-Q3_K_S.gguf filter=lfs diff=lfs merge=lfs -text
EXAONE-3.5-2.4B-Instruct-IQ3_XS.gguf filter=lfs diff=lfs merge=lfs -text
EXAONE-3.5-2.4B-Instruct-Q2_K_L.gguf filter=lfs diff=lfs merge=lfs -text
EXAONE-3.5-2.4B-Instruct-Q2_K.gguf filter=lfs diff=lfs merge=lfs -text
EXAONE-3.5-2.4B-Instruct-IQ2_M.gguf filter=lfs diff=lfs merge=lfs -text
EXAONE-3.5-2.4B-Instruct-f32.gguf filter=lfs diff=lfs merge=lfs -text
EXAONE-3.5-2.4B-Instruct.imatrix filter=lfs diff=lfs merge=lfs -text

3
EXAONE-3.5-2.4B-Instruct-IQ2_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:448a7b9ec17e95004d92d0c0199a5c8cf7911559e3ee8e8457e15bb11341ceab
size 1023555616

3
EXAONE-3.5-2.4B-Instruct-IQ3_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:5ada23941abccc9be7baf928d6babc8afacb8bd283e3b2e31008d4881c823cf3
size 1293287456

3
EXAONE-3.5-2.4B-Instruct-IQ3_XS.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:6c7af3ea7cb7d9d1fc334b518194bd3b7879592950d834849347ece024e8f99e
size 1208776736

3
EXAONE-3.5-2.4B-Instruct-IQ4_NL.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:ea78c049c9956b04c1c1f98dbcebfcf22a01663651ae0f93e7f68f9f94fff8a6
size 1578916896

3
EXAONE-3.5-2.4B-Instruct-IQ4_XS.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:ff639f1a43481d92b998c62d7b6d648b8f07a70372fb7f945236fa1025122f56
size 1505291296

3
EXAONE-3.5-2.4B-Instruct-Q2_K.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:47b91bfe502dff11af078c5c7b90f146b1a9d8a4d885f035f52d757997440dca
size 1096459296

3
EXAONE-3.5-2.4B-Instruct-Q2_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:6c5c172bec6aaa63fbb123d215b7b6066f8865bd1484da3999150c1939b0704d
size 1352459296

3
EXAONE-3.5-2.4B-Instruct-Q3_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:7380e5be6431cc81c151f9b5f13848bd12227f845094013beaa0257e7abb940a
size 1458622496

3
EXAONE-3.5-2.4B-Instruct-Q3_K_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:5d701ae13dd866e6de316aab581a561395c36b136926c76c281107b415e78d22
size 1361793056

3
EXAONE-3.5-2.4B-Instruct-Q3_K_S.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:29629d3d4380062d85b7e28aafec1c2b631f286bfc077fb286496aa8042237af
size 1253336096

3
EXAONE-3.5-2.4B-Instruct-Q3_K_XL.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:7aaf0eeb8e7b8e9f459886d72baf425b821222b261141f5dcbfa07a308608387
size 1687998496

3
EXAONE-3.5-2.4B-Instruct-Q4_0.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:970b49396842ab4a1fc85ed14eb55e7d09717bcb3c5b82b60ad5bbdc85503a19
size 1576213536

3
EXAONE-3.5-2.4B-Instruct-Q4_0_4_4.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:53f904f7caee1b8d26160c4e2acdcadb5e6e910e1541e793864cbd1bc8935a6a
size 1572772896

3
EXAONE-3.5-2.4B-Instruct-Q4_0_4_8.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:f0595a9347c96d382c83f1f3cdb3b1bc28b704075e90777fc12603117332e65f
size 1572772896

3
EXAONE-3.5-2.4B-Instruct-Q4_0_8_8.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:7b0f8deea495fa13fa27199534057f9ac5a46b5129b2f52a92216bf5c8e6af7f
size 1572772896

3
EXAONE-3.5-2.4B-Instruct-Q4_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:e88acbd6bb66146f25a8a0c73f7538560badbd36279cb2f41704ba16b4827a66
size 1839478816

3
EXAONE-3.5-2.4B-Instruct-Q4_K_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:d697402c44f2cadcd55c64f4a244c1852de764caa55cb7347904be97994f87ae
size 1644918816

3
EXAONE-3.5-2.4B-Instruct-Q4_K_S.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:15a8ffb76edeb8254cf02483c2a785811dbf0f15fa6db0925001745a37e2f9a5
size 1580473376

3
EXAONE-3.5-2.4B-Instruct-Q5_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:fbdd9edd57d55931d75ca5ea38f45d13211d529d6c8e4101ff1ad5d2206c00ad
size 2072377376

3
EXAONE-3.5-2.4B-Instruct-Q5_K_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:e62580560d6c6cb8dc6be5e943844bb7937589f21727cce284cc8323d60baaac
size 1910585376

3
EXAONE-3.5-2.4B-Instruct-Q5_K_S.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:f656b1690090495e35b990c78668946b091b265a8ff1492ffb564befe702339e
size 1873419296

3
EXAONE-3.5-2.4B-Instruct-Q6_K.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:c472ded6067fab940c05dd7efbe65093d236144491ddfefe026ee6f2a4017e7e
size 2192856096

3
EXAONE-3.5-2.4B-Instruct-Q6_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:160cabfb4a68192e46fbdb16497a8a6d5cf828868cc88dbc548a46314002a949
size 2319832096

3
EXAONE-3.5-2.4B-Instruct-Q8_0.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:3bd735d9304e61e45a69ce513f1c4fa3b88f384f6aedcfc74e52d6ce60cb78a8
size 2838846496

3
EXAONE-3.5-2.4B-Instruct-f16.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:cf3134886416a080792442d2d76c57376bec649a6f28e1e1d7da5eb53602e8af
size 5339454496

3
EXAONE-3.5-2.4B-Instruct-f32.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:c25bf318e0e82d1f078908f65eacff7681e8aefc8f5f77b24a45b6235f192800
size 10674084640

3
EXAONE-3.5-2.4B-Instruct.imatrix Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:50c26e497302a9a250ba50cc639de4051e17029bdb0783c7bd1992c1891805e3
size 2710328

178
README.md Normal file
View File

@@ -0,0 +1,178 @@
---
quantized_by: bartowski
pipeline_tag: text-generation
license_name: exaone
language:
- en
- ko
base_model: LGAI-EXAONE/EXAONE-3.5-2.4B-Instruct
tags:
- lg-ai
- exaone
- exaone-3.5
license_link: LICENSE
license: other
---
## Llamacpp imatrix Quantizations of EXAONE-3.5-2.4B-Instruct
Using <a href="https://github.com/ggerganov/llama.cpp/">llama.cpp</a> release <a href="https://github.com/ggerganov/llama.cpp/releases/tag/b4273">b4273</a> for quantization.
Original model: https://huggingface.co/LGAI-EXAONE/EXAONE-3.5-2.4B-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
```
[|system|]{system_prompt}[|endofturn|]
[|user|]{prompt}
[|assistant|]
```
## Download a file (not the whole branch) from below:
| Filename | Quant type | File Size | Split | Description |
| -------- | ---------- | --------- | ----- | ----------- |
| [EXAONE-3.5-2.4B-Instruct-f32.gguf](https://huggingface.co/bartowski/EXAONE-3.5-2.4B-Instruct-GGUF/blob/main/EXAONE-3.5-2.4B-Instruct-f32.gguf) | f32 | 10.67GB | false | Full F32 weights. |
| [EXAONE-3.5-2.4B-Instruct-f16.gguf](https://huggingface.co/bartowski/EXAONE-3.5-2.4B-Instruct-GGUF/blob/main/EXAONE-3.5-2.4B-Instruct-f16.gguf) | f16 | 5.34GB | false | Full F16 weights. |
| [EXAONE-3.5-2.4B-Instruct-Q8_0.gguf](https://huggingface.co/bartowski/EXAONE-3.5-2.4B-Instruct-GGUF/blob/main/EXAONE-3.5-2.4B-Instruct-Q8_0.gguf) | Q8_0 | 2.84GB | false | Extremely high quality, generally unneeded but max available quant. |
| [EXAONE-3.5-2.4B-Instruct-Q6_K_L.gguf](https://huggingface.co/bartowski/EXAONE-3.5-2.4B-Instruct-GGUF/blob/main/EXAONE-3.5-2.4B-Instruct-Q6_K_L.gguf) | Q6_K_L | 2.32GB | false | Uses Q8_0 for embed and output weights. Very high quality, near perfect, *recommended*. |
| [EXAONE-3.5-2.4B-Instruct-Q6_K.gguf](https://huggingface.co/bartowski/EXAONE-3.5-2.4B-Instruct-GGUF/blob/main/EXAONE-3.5-2.4B-Instruct-Q6_K.gguf) | Q6_K | 2.19GB | false | Very high quality, near perfect, *recommended*. |
| [EXAONE-3.5-2.4B-Instruct-Q5_K_L.gguf](https://huggingface.co/bartowski/EXAONE-3.5-2.4B-Instruct-GGUF/blob/main/EXAONE-3.5-2.4B-Instruct-Q5_K_L.gguf) | Q5_K_L | 2.07GB | false | Uses Q8_0 for embed and output weights. High quality, *recommended*. |
| [EXAONE-3.5-2.4B-Instruct-Q5_K_M.gguf](https://huggingface.co/bartowski/EXAONE-3.5-2.4B-Instruct-GGUF/blob/main/EXAONE-3.5-2.4B-Instruct-Q5_K_M.gguf) | Q5_K_M | 1.91GB | false | High quality, *recommended*. |
| [EXAONE-3.5-2.4B-Instruct-Q5_K_S.gguf](https://huggingface.co/bartowski/EXAONE-3.5-2.4B-Instruct-GGUF/blob/main/EXAONE-3.5-2.4B-Instruct-Q5_K_S.gguf) | Q5_K_S | 1.87GB | false | High quality, *recommended*. |
| [EXAONE-3.5-2.4B-Instruct-Q4_K_L.gguf](https://huggingface.co/bartowski/EXAONE-3.5-2.4B-Instruct-GGUF/blob/main/EXAONE-3.5-2.4B-Instruct-Q4_K_L.gguf) | Q4_K_L | 1.84GB | false | Uses Q8_0 for embed and output weights. Good quality, *recommended*. |
| [EXAONE-3.5-2.4B-Instruct-Q3_K_XL.gguf](https://huggingface.co/bartowski/EXAONE-3.5-2.4B-Instruct-GGUF/blob/main/EXAONE-3.5-2.4B-Instruct-Q3_K_XL.gguf) | Q3_K_XL | 1.69GB | false | Uses Q8_0 for embed and output weights. Lower quality but usable, good for low RAM availability. |
| [EXAONE-3.5-2.4B-Instruct-Q4_K_M.gguf](https://huggingface.co/bartowski/EXAONE-3.5-2.4B-Instruct-GGUF/blob/main/EXAONE-3.5-2.4B-Instruct-Q4_K_M.gguf) | Q4_K_M | 1.64GB | false | Good quality, default size for most use cases, *recommended*. |
| [EXAONE-3.5-2.4B-Instruct-Q4_K_S.gguf](https://huggingface.co/bartowski/EXAONE-3.5-2.4B-Instruct-GGUF/blob/main/EXAONE-3.5-2.4B-Instruct-Q4_K_S.gguf) | Q4_K_S | 1.58GB | false | Slightly lower quality with more space savings, *recommended*. |
| [EXAONE-3.5-2.4B-Instruct-Q4_0.gguf](https://huggingface.co/bartowski/EXAONE-3.5-2.4B-Instruct-GGUF/blob/main/EXAONE-3.5-2.4B-Instruct-Q4_0.gguf) | Q4_0 | 1.58GB | false | Legacy format, offers online repacking for ARM CPU inference. |
| [EXAONE-3.5-2.4B-Instruct-IQ4_NL.gguf](https://huggingface.co/bartowski/EXAONE-3.5-2.4B-Instruct-GGUF/blob/main/EXAONE-3.5-2.4B-Instruct-IQ4_NL.gguf) | IQ4_NL | 1.58GB | false | Similar to IQ4_XS, but slightly larger. Offers online repacking for ARM CPU inference. |
| [EXAONE-3.5-2.4B-Instruct-Q4_0_8_8.gguf](https://huggingface.co/bartowski/EXAONE-3.5-2.4B-Instruct-GGUF/blob/main/EXAONE-3.5-2.4B-Instruct-Q4_0_8_8.gguf) | Q4_0_8_8 | 1.57GB | false | Optimized for ARM and AVX inference. Requires 'sve' support for ARM (see details below). *Don't use on Mac*. |
| [EXAONE-3.5-2.4B-Instruct-Q4_0_4_8.gguf](https://huggingface.co/bartowski/EXAONE-3.5-2.4B-Instruct-GGUF/blob/main/EXAONE-3.5-2.4B-Instruct-Q4_0_4_8.gguf) | Q4_0_4_8 | 1.57GB | false | Optimized for ARM inference. Requires 'i8mm' support (see details below). *Don't use on Mac*. |
| [EXAONE-3.5-2.4B-Instruct-Q4_0_4_4.gguf](https://huggingface.co/bartowski/EXAONE-3.5-2.4B-Instruct-GGUF/blob/main/EXAONE-3.5-2.4B-Instruct-Q4_0_4_4.gguf) | Q4_0_4_4 | 1.57GB | false | Optimized for ARM inference. Should work well on all ARM chips, not for use with GPUs. *Don't use on Mac*. |
| [EXAONE-3.5-2.4B-Instruct-IQ4_XS.gguf](https://huggingface.co/bartowski/EXAONE-3.5-2.4B-Instruct-GGUF/blob/main/EXAONE-3.5-2.4B-Instruct-IQ4_XS.gguf) | IQ4_XS | 1.51GB | false | Decent quality, smaller than Q4_K_S with similar performance, *recommended*. |
| [EXAONE-3.5-2.4B-Instruct-Q3_K_L.gguf](https://huggingface.co/bartowski/EXAONE-3.5-2.4B-Instruct-GGUF/blob/main/EXAONE-3.5-2.4B-Instruct-Q3_K_L.gguf) | Q3_K_L | 1.46GB | false | Lower quality but usable, good for low RAM availability. |
| [EXAONE-3.5-2.4B-Instruct-Q3_K_M.gguf](https://huggingface.co/bartowski/EXAONE-3.5-2.4B-Instruct-GGUF/blob/main/EXAONE-3.5-2.4B-Instruct-Q3_K_M.gguf) | Q3_K_M | 1.36GB | false | Low quality. |
| [EXAONE-3.5-2.4B-Instruct-Q2_K_L.gguf](https://huggingface.co/bartowski/EXAONE-3.5-2.4B-Instruct-GGUF/blob/main/EXAONE-3.5-2.4B-Instruct-Q2_K_L.gguf) | Q2_K_L | 1.35GB | false | Uses Q8_0 for embed and output weights. Very low quality but surprisingly usable. |
| [EXAONE-3.5-2.4B-Instruct-IQ3_M.gguf](https://huggingface.co/bartowski/EXAONE-3.5-2.4B-Instruct-GGUF/blob/main/EXAONE-3.5-2.4B-Instruct-IQ3_M.gguf) | IQ3_M | 1.29GB | false | Medium-low quality, new method with decent performance comparable to Q3_K_M. |
| [EXAONE-3.5-2.4B-Instruct-Q3_K_S.gguf](https://huggingface.co/bartowski/EXAONE-3.5-2.4B-Instruct-GGUF/blob/main/EXAONE-3.5-2.4B-Instruct-Q3_K_S.gguf) | Q3_K_S | 1.25GB | false | Low quality, not recommended. |
| [EXAONE-3.5-2.4B-Instruct-IQ3_XS.gguf](https://huggingface.co/bartowski/EXAONE-3.5-2.4B-Instruct-GGUF/blob/main/EXAONE-3.5-2.4B-Instruct-IQ3_XS.gguf) | IQ3_XS | 1.21GB | false | Lower quality, new method with decent performance, slightly better than Q3_K_S. |
| [EXAONE-3.5-2.4B-Instruct-Q2_K.gguf](https://huggingface.co/bartowski/EXAONE-3.5-2.4B-Instruct-GGUF/blob/main/EXAONE-3.5-2.4B-Instruct-Q2_K.gguf) | Q2_K | 1.10GB | false | Very low quality but surprisingly usable. |
| [EXAONE-3.5-2.4B-Instruct-IQ2_M.gguf](https://huggingface.co/bartowski/EXAONE-3.5-2.4B-Instruct-GGUF/blob/main/EXAONE-3.5-2.4B-Instruct-IQ2_M.gguf) | IQ2_M | 1.02GB | 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/EXAONE-3.5-2.4B-Instruct-GGUF --include "EXAONE-3.5-2.4B-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/EXAONE-3.5-2.4B-Instruct-GGUF --include "EXAONE-3.5-2.4B-Instruct-Q8_0/*" --local-dir ./
```
You can either specify a new local-dir (EXAONE-3.5-2.4B-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}