bartowski/PowerInfer_SmallThinker-4BA0.6B-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/PowerInfer_SmallThinker-4BA0.6B-Instruct-GGUF
Source: Original Platform
This commit is contained in:
ModelHub XC
2026-10-12 01:54:13 +08:00
commit 76be5b268d
28 changed files with 293 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

47
.gitattributes vendored Normal file
View File

@@ -0,0 +1,47 @@
*.7z filter=lfs diff=lfs merge=lfs -text
*.arrow filter=lfs diff=lfs merge=lfs -text
*.bin filter=lfs diff=lfs merge=lfs -text
*.bin.* filter=lfs diff=lfs merge=lfs -text
*.bz2 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
*.model filter=lfs diff=lfs merge=lfs -text
*.msgpack 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
*.pt filter=lfs diff=lfs merge=lfs -text
*.pth filter=lfs diff=lfs merge=lfs -text
*.rar filter=lfs diff=lfs merge=lfs -text
saved_model/**/* 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
*.xz filter=lfs diff=lfs merge=lfs -text
*.zip filter=lfs diff=lfs merge=lfs -text
*.zstandard filter=lfs diff=lfs merge=lfs -text
*.tfevents* filter=lfs diff=lfs merge=lfs -text
*.db* filter=lfs diff=lfs merge=lfs -text
*.ark* filter=lfs diff=lfs merge=lfs -text
**/*ckpt*data* filter=lfs diff=lfs merge=lfs -text
**/*ckpt*.meta filter=lfs diff=lfs merge=lfs -text
**/*ckpt*.index filter=lfs diff=lfs merge=lfs -text
*.safetensors filter=lfs diff=lfs merge=lfs -text
*.ckpt filter=lfs diff=lfs merge=lfs -text
*.gguf* filter=lfs diff=lfs merge=lfs -text
*.ggml filter=lfs diff=lfs merge=lfs -text
*.llamafile* filter=lfs diff=lfs merge=lfs -text
*.pt2 filter=lfs diff=lfs merge=lfs -text
*.mlmodel filter=lfs diff=lfs merge=lfs -text
*.npy filter=lfs diff=lfs merge=lfs -text
*.npz filter=lfs diff=lfs merge=lfs -text
*.pickle filter=lfs diff=lfs merge=lfs -text
*.pkl filter=lfs diff=lfs merge=lfs -text
*.tar filter=lfs diff=lfs merge=lfs -text
*.wasm filter=lfs diff=lfs merge=lfs -text
*.zst filter=lfs diff=lfs merge=lfs -text
*tfevents* filter=lfs diff=lfs merge=lfs -text

3
PowerInfer_SmallThinker-4BA0.6B-Instruct-IQ2_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:eeff79b9f9f5b34eefe42a4cde5fdab5d08857138dc78bacb47ae07fb369167a
size 1485453952

3
PowerInfer_SmallThinker-4BA0.6B-Instruct-IQ3_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:91fdee23431447b4ff863873714d3e74e75b28b661179886979a53e387789a55
size 2046504064

3
PowerInfer_SmallThinker-4BA0.6B-Instruct-IQ3_XS.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:99903b3174b392d442543017e78ac9e1c4ce4cd216a2f409afb646a961cb8aae
size 1877814400

3
PowerInfer_SmallThinker-4BA0.6B-Instruct-IQ3_XXS.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:ab1ba44844a0933b3851cd2aae2cca24f94bb41de5fc03371d2b0189ee47654a
size 1785477760

3
PowerInfer_SmallThinker-4BA0.6B-Instruct-IQ4_NL.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:1326ee4f4c111729938fb30dc40e9f81778d624261aed5aa8d4caac2e29dfe5f
size 2490179200

3
PowerInfer_SmallThinker-4BA0.6B-Instruct-IQ4_XS.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:710f929a8fffc09701d4006eacfd9ebe3cbf557f5559e820a2bd626027398a23
size 2367280768

3
PowerInfer_SmallThinker-4BA0.6B-Instruct-Q2_K.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:66f6d45bda665d3b9be71212893dd8d3ecf888d4bbc97b9e6fb9eecbd4cd1cf4
size 1628209792

3
PowerInfer_SmallThinker-4BA0.6B-Instruct-Q2_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:2e5b0cce554aaaff058c791db38a990a1ab9ccd3baf89789d56ef9edb4c8c9a6
size 1856113792

3
PowerInfer_SmallThinker-4BA0.6B-Instruct-Q3_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:c9ed297adc05656e6563cb86e60e9ebb25c20b592575142f5593a4126b54c4fc
size 2104110208

3
PowerInfer_SmallThinker-4BA0.6B-Instruct-Q3_K_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:dd5ca140fdd8b2fb1775d6ccf1ef9a69ba6fca0750827e46f4da5eee34329900
size 2046602368

3
PowerInfer_SmallThinker-4BA0.6B-Instruct-Q3_K_S.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:9f1b24722e9be59b685bdc56452b2e22fc4c1a45ea17603e3c8242c7b9bde5cd
size 1964616832

3
PowerInfer_SmallThinker-4BA0.6B-Instruct-Q3_K_XL.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:8fbad32fc7a176c700c5f0453dcbf3f4985ebf3ef01290e48066873592ae1e51
size 2308312192

3
PowerInfer_SmallThinker-4BA0.6B-Instruct-Q4_0.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:563c99cc38f56d1954b3aa5b992248001a14a08bf3c2bdac33648c34592dbe69
size 2516917888

3
PowerInfer_SmallThinker-4BA0.6B-Instruct-Q4_1.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:e02e58508f387729a2178295951396bd69cc418afdf3b8f6a61c234c92737ec7
size 2730471040

3
PowerInfer_SmallThinker-4BA0.6B-Instruct-Q4_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:90d75848333a4b6bd7d84490ef85fe4b16a96cdb96a6cfb791b5fecdafc39c78
size 2819148928

3
PowerInfer_SmallThinker-4BA0.6B-Instruct-Q4_K_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:f1ee3defe19e937f3d4994aefe22b835cecf61f15ad16b9f00f28c3f64ed7b0a
size 2645941888

3
PowerInfer_SmallThinker-4BA0.6B-Instruct-Q4_K_S.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:432ab87caf413fc389bd53b6037c42aa9e2c3a06e723a86bfec560be5e35f62e
size 2564300416

3
PowerInfer_SmallThinker-4BA0.6B-Instruct-Q5_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:017f061a5b0afd9823b1944e8043ba32f3160979753d6ecb1290252b4eee6592
size 3206859904

3
PowerInfer_SmallThinker-4BA0.6B-Instruct-Q5_K_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:f10d64512b9946cb2a03f114d1719b6397b1861460d2b4e823b64737ed5fe454
size 3062824576

3
PowerInfer_SmallThinker-4BA0.6B-Instruct-Q5_K_S.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:2b3ec5a6328adcada753dfcc0e043db5dbb61c2face6a2a9b48b4babdfbf49dc
size 2981772928

3
PowerInfer_SmallThinker-4BA0.6B-Instruct-Q6_K.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:cd1978e5529016ff2769ecbf9ee317ab0eb5e89fe417bd78f7e7b721a87cfb0c
size 3515789440

3
PowerInfer_SmallThinker-4BA0.6B-Instruct-Q6_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:c1f2a1d9c7bce28facafc1fd2fb309c0f971a764b0015360b49eb1691ee16545
size 3628829824

3
PowerInfer_SmallThinker-4BA0.6B-Instruct-Q8_0.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:9bb022d5229f2de1a86045e75bda113fc5fd295a56f199ddc55c96eea84a26f5
size 4546104448

3
PowerInfer_SmallThinker-4BA0.6B-Instruct-bf16.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:872aec935914d9391bd89dc0fd3d2677d2e4128c63b8b142e134c0f9d2b430c7
size 8546216768

3
PowerInfer_SmallThinker-4BA0.6B-Instruct-imatrix.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:0df2a9f44764191880546437c355b1c7dc53fec91be9885bbf88fe611ebb3388
size 16763072

170
README.md Normal file
View File

@@ -0,0 +1,170 @@
---
quantized_by: bartowski
pipeline_tag: text-generation
base_model_relation: quantized
base_model: PowerInfer/SmallThinker-4BA0.6B-Instruct
---
## Llamacpp imatrix Quantizations of SmallThinker-4BA0.6B-Instruct by PowerInfer
Using <a href="https://github.com/ggerganov/llama.cpp/">llama.cpp</a> release <a href="https://github.com/ggerganov/llama.cpp/releases/tag/b6012">b6012</a> for quantization.
Original model: https://huggingface.co/PowerInfer/SmallThinker-4BA0.6B-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/)
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 |
| -------- | ---------- | --------- | ----- | ----------- |
| [SmallThinker-4BA0.6B-Instruct-bf16.gguf](https://huggingface.co/bartowski/PowerInfer_SmallThinker-4BA0.6B-Instruct-GGUF/blob/main/PowerInfer_SmallThinker-4BA0.6B-Instruct-bf16.gguf) | bf16 | 8.55GB | false | Full BF16 weights. |
| [SmallThinker-4BA0.6B-Instruct-Q8_0.gguf](https://huggingface.co/bartowski/PowerInfer_SmallThinker-4BA0.6B-Instruct-GGUF/blob/main/PowerInfer_SmallThinker-4BA0.6B-Instruct-Q8_0.gguf) | Q8_0 | 4.55GB | false | Extremely high quality, generally unneeded but max available quant. |
| [SmallThinker-4BA0.6B-Instruct-Q6_K_L.gguf](https://huggingface.co/bartowski/PowerInfer_SmallThinker-4BA0.6B-Instruct-GGUF/blob/main/PowerInfer_SmallThinker-4BA0.6B-Instruct-Q6_K_L.gguf) | Q6_K_L | 3.63GB | false | Uses Q8_0 for embed and output weights. Very high quality, near perfect, *recommended*. |
| [SmallThinker-4BA0.6B-Instruct-Q6_K.gguf](https://huggingface.co/bartowski/PowerInfer_SmallThinker-4BA0.6B-Instruct-GGUF/blob/main/PowerInfer_SmallThinker-4BA0.6B-Instruct-Q6_K.gguf) | Q6_K | 3.52GB | false | Very high quality, near perfect, *recommended*. |
| [SmallThinker-4BA0.6B-Instruct-Q5_K_L.gguf](https://huggingface.co/bartowski/PowerInfer_SmallThinker-4BA0.6B-Instruct-GGUF/blob/main/PowerInfer_SmallThinker-4BA0.6B-Instruct-Q5_K_L.gguf) | Q5_K_L | 3.21GB | false | Uses Q8_0 for embed and output weights. High quality, *recommended*. |
| [SmallThinker-4BA0.6B-Instruct-Q5_K_M.gguf](https://huggingface.co/bartowski/PowerInfer_SmallThinker-4BA0.6B-Instruct-GGUF/blob/main/PowerInfer_SmallThinker-4BA0.6B-Instruct-Q5_K_M.gguf) | Q5_K_M | 3.06GB | false | High quality, *recommended*. |
| [SmallThinker-4BA0.6B-Instruct-Q5_K_S.gguf](https://huggingface.co/bartowski/PowerInfer_SmallThinker-4BA0.6B-Instruct-GGUF/blob/main/PowerInfer_SmallThinker-4BA0.6B-Instruct-Q5_K_S.gguf) | Q5_K_S | 2.98GB | false | High quality, *recommended*. |
| [SmallThinker-4BA0.6B-Instruct-Q4_K_L.gguf](https://huggingface.co/bartowski/PowerInfer_SmallThinker-4BA0.6B-Instruct-GGUF/blob/main/PowerInfer_SmallThinker-4BA0.6B-Instruct-Q4_K_L.gguf) | Q4_K_L | 2.82GB | false | Uses Q8_0 for embed and output weights. Good quality, *recommended*. |
| [SmallThinker-4BA0.6B-Instruct-Q4_1.gguf](https://huggingface.co/bartowski/PowerInfer_SmallThinker-4BA0.6B-Instruct-GGUF/blob/main/PowerInfer_SmallThinker-4BA0.6B-Instruct-Q4_1.gguf) | Q4_1 | 2.73GB | false | Legacy format, similar performance to Q4_K_S but with improved tokens/watt on Apple silicon. |
| [SmallThinker-4BA0.6B-Instruct-Q4_K_M.gguf](https://huggingface.co/bartowski/PowerInfer_SmallThinker-4BA0.6B-Instruct-GGUF/blob/main/PowerInfer_SmallThinker-4BA0.6B-Instruct-Q4_K_M.gguf) | Q4_K_M | 2.65GB | false | Good quality, default size for most use cases, *recommended*. |
| [SmallThinker-4BA0.6B-Instruct-Q4_K_S.gguf](https://huggingface.co/bartowski/PowerInfer_SmallThinker-4BA0.6B-Instruct-GGUF/blob/main/PowerInfer_SmallThinker-4BA0.6B-Instruct-Q4_K_S.gguf) | Q4_K_S | 2.56GB | false | Slightly lower quality with more space savings, *recommended*. |
| [SmallThinker-4BA0.6B-Instruct-Q4_0.gguf](https://huggingface.co/bartowski/PowerInfer_SmallThinker-4BA0.6B-Instruct-GGUF/blob/main/PowerInfer_SmallThinker-4BA0.6B-Instruct-Q4_0.gguf) | Q4_0 | 2.52GB | false | Legacy format, offers online repacking for ARM and AVX CPU inference. |
| [SmallThinker-4BA0.6B-Instruct-IQ4_NL.gguf](https://huggingface.co/bartowski/PowerInfer_SmallThinker-4BA0.6B-Instruct-GGUF/blob/main/PowerInfer_SmallThinker-4BA0.6B-Instruct-IQ4_NL.gguf) | IQ4_NL | 2.49GB | false | Similar to IQ4_XS, but slightly larger. Offers online repacking for ARM CPU inference. |
| [SmallThinker-4BA0.6B-Instruct-IQ4_XS.gguf](https://huggingface.co/bartowski/PowerInfer_SmallThinker-4BA0.6B-Instruct-GGUF/blob/main/PowerInfer_SmallThinker-4BA0.6B-Instruct-IQ4_XS.gguf) | IQ4_XS | 2.37GB | false | Decent quality, smaller than Q4_K_S with similar performance, *recommended*. |
| [SmallThinker-4BA0.6B-Instruct-Q3_K_XL.gguf](https://huggingface.co/bartowski/PowerInfer_SmallThinker-4BA0.6B-Instruct-GGUF/blob/main/PowerInfer_SmallThinker-4BA0.6B-Instruct-Q3_K_XL.gguf) | Q3_K_XL | 2.31GB | false | Uses Q8_0 for embed and output weights. Lower quality but usable, good for low RAM availability. |
| [SmallThinker-4BA0.6B-Instruct-Q3_K_L.gguf](https://huggingface.co/bartowski/PowerInfer_SmallThinker-4BA0.6B-Instruct-GGUF/blob/main/PowerInfer_SmallThinker-4BA0.6B-Instruct-Q3_K_L.gguf) | Q3_K_L | 2.10GB | false | Lower quality but usable, good for low RAM availability. |
| [SmallThinker-4BA0.6B-Instruct-Q3_K_M.gguf](https://huggingface.co/bartowski/PowerInfer_SmallThinker-4BA0.6B-Instruct-GGUF/blob/main/PowerInfer_SmallThinker-4BA0.6B-Instruct-Q3_K_M.gguf) | Q3_K_M | 2.05GB | false | Low quality. |
| [SmallThinker-4BA0.6B-Instruct-IQ3_M.gguf](https://huggingface.co/bartowski/PowerInfer_SmallThinker-4BA0.6B-Instruct-GGUF/blob/main/PowerInfer_SmallThinker-4BA0.6B-Instruct-IQ3_M.gguf) | IQ3_M | 2.05GB | false | Medium-low quality, new method with decent performance comparable to Q3_K_M. |
| [SmallThinker-4BA0.6B-Instruct-Q3_K_S.gguf](https://huggingface.co/bartowski/PowerInfer_SmallThinker-4BA0.6B-Instruct-GGUF/blob/main/PowerInfer_SmallThinker-4BA0.6B-Instruct-Q3_K_S.gguf) | Q3_K_S | 1.96GB | false | Low quality, not recommended. |
| [SmallThinker-4BA0.6B-Instruct-IQ3_XS.gguf](https://huggingface.co/bartowski/PowerInfer_SmallThinker-4BA0.6B-Instruct-GGUF/blob/main/PowerInfer_SmallThinker-4BA0.6B-Instruct-IQ3_XS.gguf) | IQ3_XS | 1.88GB | false | Lower quality, new method with decent performance, slightly better than Q3_K_S. |
| [SmallThinker-4BA0.6B-Instruct-Q2_K_L.gguf](https://huggingface.co/bartowski/PowerInfer_SmallThinker-4BA0.6B-Instruct-GGUF/blob/main/PowerInfer_SmallThinker-4BA0.6B-Instruct-Q2_K_L.gguf) | Q2_K_L | 1.86GB | false | Uses Q8_0 for embed and output weights. Very low quality but surprisingly usable. |
| [SmallThinker-4BA0.6B-Instruct-IQ3_XXS.gguf](https://huggingface.co/bartowski/PowerInfer_SmallThinker-4BA0.6B-Instruct-GGUF/blob/main/PowerInfer_SmallThinker-4BA0.6B-Instruct-IQ3_XXS.gguf) | IQ3_XXS | 1.79GB | false | Lower quality, new method with decent performance, comparable to Q3 quants. |
| [SmallThinker-4BA0.6B-Instruct-Q2_K.gguf](https://huggingface.co/bartowski/PowerInfer_SmallThinker-4BA0.6B-Instruct-GGUF/blob/main/PowerInfer_SmallThinker-4BA0.6B-Instruct-Q2_K.gguf) | Q2_K | 1.63GB | false | Very low quality but surprisingly usable. |
| [SmallThinker-4BA0.6B-Instruct-IQ2_M.gguf](https://huggingface.co/bartowski/PowerInfer_SmallThinker-4BA0.6B-Instruct-GGUF/blob/main/PowerInfer_SmallThinker-4BA0.6B-Instruct-IQ2_M.gguf) | IQ2_M | 1.49GB | 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/PowerInfer_SmallThinker-4BA0.6B-Instruct-GGUF --include "PowerInfer_SmallThinker-4BA0.6B-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/PowerInfer_SmallThinker-4BA0.6B-Instruct-GGUF --include "PowerInfer_SmallThinker-4BA0.6B-Instruct-Q8_0/*" --local-dir ./
```
You can either specify a new local-dir (PowerInfer_SmallThinker-4BA0.6B-Instruct-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, 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

1
configuration.json Normal file
View File

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