bartowski/GWQ-9B-Preview2-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/GWQ-9B-Preview2-GGUF
Source: Original Platform
This commit is contained in:
ModelHub XC
2026-09-18 12:30:12 +08:00
commit d8bf4da518
27 changed files with 307 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

59
.gitattributes vendored Normal file
View File

@@ -0,0 +1,59 @@
*.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
GWQ-9B-Preview2-Q8_0.gguf filter=lfs diff=lfs merge=lfs -text
GWQ-9B-Preview2-Q6_K_L.gguf filter=lfs diff=lfs merge=lfs -text
GWQ-9B-Preview2-Q6_K.gguf filter=lfs diff=lfs merge=lfs -text
GWQ-9B-Preview2-Q5_K_L.gguf filter=lfs diff=lfs merge=lfs -text
GWQ-9B-Preview2-Q5_K_M.gguf filter=lfs diff=lfs merge=lfs -text
GWQ-9B-Preview2-Q5_K_S.gguf filter=lfs diff=lfs merge=lfs -text
GWQ-9B-Preview2-Q4_K_L.gguf filter=lfs diff=lfs merge=lfs -text
GWQ-9B-Preview2-Q4_K_M.gguf filter=lfs diff=lfs merge=lfs -text
GWQ-9B-Preview2-Q4_K_S.gguf filter=lfs diff=lfs merge=lfs -text
GWQ-9B-Preview2-Q4_1.gguf filter=lfs diff=lfs merge=lfs -text
GWQ-9B-Preview2-Q4_0.gguf filter=lfs diff=lfs merge=lfs -text
GWQ-9B-Preview2-IQ4_NL.gguf filter=lfs diff=lfs merge=lfs -text
GWQ-9B-Preview2-IQ4_XS.gguf filter=lfs diff=lfs merge=lfs -text
GWQ-9B-Preview2-Q3_K_XL.gguf filter=lfs diff=lfs merge=lfs -text
GWQ-9B-Preview2-Q3_K_L.gguf filter=lfs diff=lfs merge=lfs -text
GWQ-9B-Preview2-Q3_K_M.gguf filter=lfs diff=lfs merge=lfs -text
GWQ-9B-Preview2-IQ3_M.gguf filter=lfs diff=lfs merge=lfs -text
GWQ-9B-Preview2-Q3_K_S.gguf filter=lfs diff=lfs merge=lfs -text
GWQ-9B-Preview2-IQ3_XS.gguf filter=lfs diff=lfs merge=lfs -text
GWQ-9B-Preview2-Q2_K_L.gguf filter=lfs diff=lfs merge=lfs -text
GWQ-9B-Preview2-Q2_K.gguf filter=lfs diff=lfs merge=lfs -text
GWQ-9B-Preview2-IQ2_M.gguf filter=lfs diff=lfs merge=lfs -text
GWQ-9B-Preview2-f16.gguf filter=lfs diff=lfs merge=lfs -text
GWQ-9B-Preview2.imatrix filter=lfs diff=lfs merge=lfs -text

3
GWQ-9B-Preview2-IQ2_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:3a60aab339397a8d687a214c45114d11811ad22caefd6b6787b942f621e63632
size 3434669632

3
GWQ-9B-Preview2-IQ3_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:b4f0ab343ed78d4abb56d1e7541e0cbfa7d931477e1b1de44931bb418acd2673
size 4494616128

3
GWQ-9B-Preview2-IQ3_XS.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:44797c970acb437aea92b62f882a702a3d3c2af2a404efb610a42a0dbcd815bc
size 4144989760

3
GWQ-9B-Preview2-IQ4_NL.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:3f30d56286cbd0375679ea7bbbebfb9b6229c9d8bae9475bc28c9e2a80ac26e5
size 5443143232

3
GWQ-9B-Preview2-IQ4_XS.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:8e2e8c415599f87618770e2b6412f2770559a15749b7788bfb186cd005ca086d
size 5183030848

3
GWQ-9B-Preview2-Q2_K.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:fe5fb56d99ea6acb2b15022e21bb4cb59d01a4fa24c5a14d894196f7269bae56
size 3805398592

3
GWQ-9B-Preview2-Q2_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:f8e716ac0ce0447560ee282a081014000afd2f2c6a892c430b50cf27f1221abb
size 4027606592

3
GWQ-9B-Preview2-Q3_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:a056d406ee0419d1ecdc16a20d34552fd25a0051563955510162e86fed85bcd0
size 5132453440

3
GWQ-9B-Preview2-Q3_K_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:ba448d026ee4386c193f45bdf043d8b38aa5b56166b10f82444c742afcb5a54f
size 4761781824

3
GWQ-9B-Preview2-Q3_K_S.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:b8cb1ef9d1214c6d0d23caf7daf4b22352fe608a4358cf22ca4e262fdb36c202
size 4337665600

3
GWQ-9B-Preview2-Q3_K_XL.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:a8cb3b585331b127ec41254c633d3f0704a4c0468741af461f8c6bd169a73766
size 5354661440

3
GWQ-9B-Preview2-Q4_0.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:70fce606004c04585f066623ad967d539c9bc79d0dcf766cdecb176b22f00b5b
size 5459199552

3
GWQ-9B-Preview2-Q4_1.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:d4092e2f5ae464389aa2cd7af201bdf822e18f2712d7413a8cfa761b3e3932fb
size 5963368000

3
GWQ-9B-Preview2-Q4_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:484afa2c84731e318fec34562c24e545f9e18791fdbc18380395bfd6a9227921
size 5983266368

3
GWQ-9B-Preview2-Q4_K_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:04da51cdb17c7e51594f6daac595161a46298b48ab5e568a85e65541d10a861f
size 5761058368

3
GWQ-9B-Preview2-Q4_K_S.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:b81374c6cbc20a40684cd505cf35fa0dc82ba2c2b92a77be1ee82156d0940cad
size 5478925888

3
GWQ-9B-Preview2-Q5_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:70cffeb530fd8aa045b457aa166f42261c3264544e7e91d886c0fbd04c3afe3e
size 6869575232

3
GWQ-9B-Preview2-Q5_K_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:85b8d637efdbfe97ddee74ed7975b450ed432d8d32461fc9b974b8102e9efb98
size 6647367232

3
GWQ-9B-Preview2-Q5_K_S.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:1c40cfd30f258fef02f28aa52f9baeb843617d633936e7b2bae0ca34f1202404
size 6483592768

3
GWQ-9B-Preview2-Q6_K.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:ab7e889da927555d8fddd3784f833e9498d65da03bfde4388a11dab5d0b228c7
size 7589070400

3
GWQ-9B-Preview2-Q6_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:e9af507d4c3d7a78bc53b74c68b15eb60b59fb53e53e3b9a1482da216b31de2a
size 7811278400

3
GWQ-9B-Preview2-Q8_0.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:0e082c6e05e57e9170f173d2b419f4505ae38caf8bce45e7d357327dde059a84
size 9827149376

3
GWQ-9B-Preview2-f16.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:266629885bf73054835616151d68d99c4e054bf3bbf14c8698f2a62f83ef4513
size 18490680640

3
GWQ-9B-Preview2.imatrix Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:7f0edc4bab9cecab6f5a893db924f3fdfebd9ee43af46518aeed1c40b7867639
size 6116900

175
README.md Normal file
View File

@@ -0,0 +1,175 @@
---
quantized_by: bartowski
pipeline_tag: text-generation
tags:
- gemma2
- text-generation-inference
- f16
language:
- en
base_model: prithivMLmods/GWQ-9B-Preview2
license: creativeml-openrail-m
---
## Llamacpp imatrix Quantizations of GWQ-9B-Preview2
Using <a href="https://github.com/ggerganov/llama.cpp/">llama.cpp</a> release <a href="https://github.com/ggerganov/llama.cpp/releases/tag/b4418">b4418</a> for quantization.
Original model: https://huggingface.co/prithivMLmods/GWQ-9B-Preview2
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
```
<bos><start_of_turn>user
{prompt}<end_of_turn>
<start_of_turn>model
<end_of_turn>
<start_of_turn>model
```
Note that this model does not support a System prompt.
## Download a file (not the whole branch) from below:
| Filename | Quant type | File Size | Split | Description |
| -------- | ---------- | --------- | ----- | ----------- |
| [GWQ-9B-Preview2-f16.gguf](https://huggingface.co/bartowski/GWQ-9B-Preview2-GGUF/blob/main/GWQ-9B-Preview2-f16.gguf) | f16 | 18.49GB | false | Full F16 weights. |
| [GWQ-9B-Preview2-Q8_0.gguf](https://huggingface.co/bartowski/GWQ-9B-Preview2-GGUF/blob/main/GWQ-9B-Preview2-Q8_0.gguf) | Q8_0 | 9.83GB | false | Extremely high quality, generally unneeded but max available quant. |
| [GWQ-9B-Preview2-Q6_K_L.gguf](https://huggingface.co/bartowski/GWQ-9B-Preview2-GGUF/blob/main/GWQ-9B-Preview2-Q6_K_L.gguf) | Q6_K_L | 7.81GB | false | Uses Q8_0 for embed and output weights. Very high quality, near perfect, *recommended*. |
| [GWQ-9B-Preview2-Q6_K.gguf](https://huggingface.co/bartowski/GWQ-9B-Preview2-GGUF/blob/main/GWQ-9B-Preview2-Q6_K.gguf) | Q6_K | 7.59GB | false | Very high quality, near perfect, *recommended*. |
| [GWQ-9B-Preview2-Q5_K_L.gguf](https://huggingface.co/bartowski/GWQ-9B-Preview2-GGUF/blob/main/GWQ-9B-Preview2-Q5_K_L.gguf) | Q5_K_L | 6.87GB | false | Uses Q8_0 for embed and output weights. High quality, *recommended*. |
| [GWQ-9B-Preview2-Q5_K_M.gguf](https://huggingface.co/bartowski/GWQ-9B-Preview2-GGUF/blob/main/GWQ-9B-Preview2-Q5_K_M.gguf) | Q5_K_M | 6.65GB | false | High quality, *recommended*. |
| [GWQ-9B-Preview2-Q5_K_S.gguf](https://huggingface.co/bartowski/GWQ-9B-Preview2-GGUF/blob/main/GWQ-9B-Preview2-Q5_K_S.gguf) | Q5_K_S | 6.48GB | false | High quality, *recommended*. |
| [GWQ-9B-Preview2-Q4_K_L.gguf](https://huggingface.co/bartowski/GWQ-9B-Preview2-GGUF/blob/main/GWQ-9B-Preview2-Q4_K_L.gguf) | Q4_K_L | 5.98GB | false | Uses Q8_0 for embed and output weights. Good quality, *recommended*. |
| [GWQ-9B-Preview2-Q4_1.gguf](https://huggingface.co/bartowski/GWQ-9B-Preview2-GGUF/blob/main/GWQ-9B-Preview2-Q4_1.gguf) | Q4_1 | 5.96GB | false | Legacy format, similar performance to Q4_K_S but with improved tokens/watt on Apple silicon. |
| [GWQ-9B-Preview2-Q4_K_M.gguf](https://huggingface.co/bartowski/GWQ-9B-Preview2-GGUF/blob/main/GWQ-9B-Preview2-Q4_K_M.gguf) | Q4_K_M | 5.76GB | false | Good quality, default size for most use cases, *recommended*. |
| [GWQ-9B-Preview2-Q4_K_S.gguf](https://huggingface.co/bartowski/GWQ-9B-Preview2-GGUF/blob/main/GWQ-9B-Preview2-Q4_K_S.gguf) | Q4_K_S | 5.48GB | false | Slightly lower quality with more space savings, *recommended*. |
| [GWQ-9B-Preview2-Q4_0.gguf](https://huggingface.co/bartowski/GWQ-9B-Preview2-GGUF/blob/main/GWQ-9B-Preview2-Q4_0.gguf) | Q4_0 | 5.46GB | false | Legacy format, offers online repacking for ARM and AVX CPU inference. |
| [GWQ-9B-Preview2-IQ4_NL.gguf](https://huggingface.co/bartowski/GWQ-9B-Preview2-GGUF/blob/main/GWQ-9B-Preview2-IQ4_NL.gguf) | IQ4_NL | 5.44GB | false | Similar to IQ4_XS, but slightly larger. Offers online repacking for ARM CPU inference. |
| [GWQ-9B-Preview2-Q3_K_XL.gguf](https://huggingface.co/bartowski/GWQ-9B-Preview2-GGUF/blob/main/GWQ-9B-Preview2-Q3_K_XL.gguf) | Q3_K_XL | 5.35GB | false | Uses Q8_0 for embed and output weights. Lower quality but usable, good for low RAM availability. |
| [GWQ-9B-Preview2-IQ4_XS.gguf](https://huggingface.co/bartowski/GWQ-9B-Preview2-GGUF/blob/main/GWQ-9B-Preview2-IQ4_XS.gguf) | IQ4_XS | 5.18GB | false | Decent quality, smaller than Q4_K_S with similar performance, *recommended*. |
| [GWQ-9B-Preview2-Q3_K_L.gguf](https://huggingface.co/bartowski/GWQ-9B-Preview2-GGUF/blob/main/GWQ-9B-Preview2-Q3_K_L.gguf) | Q3_K_L | 5.13GB | false | Lower quality but usable, good for low RAM availability. |
| [GWQ-9B-Preview2-Q3_K_M.gguf](https://huggingface.co/bartowski/GWQ-9B-Preview2-GGUF/blob/main/GWQ-9B-Preview2-Q3_K_M.gguf) | Q3_K_M | 4.76GB | false | Low quality. |
| [GWQ-9B-Preview2-IQ3_M.gguf](https://huggingface.co/bartowski/GWQ-9B-Preview2-GGUF/blob/main/GWQ-9B-Preview2-IQ3_M.gguf) | IQ3_M | 4.49GB | false | Medium-low quality, new method with decent performance comparable to Q3_K_M. |
| [GWQ-9B-Preview2-Q3_K_S.gguf](https://huggingface.co/bartowski/GWQ-9B-Preview2-GGUF/blob/main/GWQ-9B-Preview2-Q3_K_S.gguf) | Q3_K_S | 4.34GB | false | Low quality, not recommended. |
| [GWQ-9B-Preview2-IQ3_XS.gguf](https://huggingface.co/bartowski/GWQ-9B-Preview2-GGUF/blob/main/GWQ-9B-Preview2-IQ3_XS.gguf) | IQ3_XS | 4.14GB | false | Lower quality, new method with decent performance, slightly better than Q3_K_S. |
| [GWQ-9B-Preview2-Q2_K_L.gguf](https://huggingface.co/bartowski/GWQ-9B-Preview2-GGUF/blob/main/GWQ-9B-Preview2-Q2_K_L.gguf) | Q2_K_L | 4.03GB | false | Uses Q8_0 for embed and output weights. Very low quality but surprisingly usable. |
| [GWQ-9B-Preview2-Q2_K.gguf](https://huggingface.co/bartowski/GWQ-9B-Preview2-GGUF/blob/main/GWQ-9B-Preview2-Q2_K.gguf) | Q2_K | 3.81GB | false | Very low quality but surprisingly usable. |
| [GWQ-9B-Preview2-IQ2_M.gguf](https://huggingface.co/bartowski/GWQ-9B-Preview2-GGUF/blob/main/GWQ-9B-Preview2-IQ2_M.gguf) | IQ2_M | 3.43GB | 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/GWQ-9B-Preview2-GGUF --include "GWQ-9B-Preview2-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/GWQ-9B-Preview2-GGUF --include "GWQ-9B-Preview2-Q8_0/*" --local-dir ./
```
You can either specify a new local-dir (GWQ-9B-Preview2-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.
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}