bartowski/Qwen_Qwen2.5-VL-7B-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/Qwen_Qwen2.5-VL-7B-Instruct-GGUF
Source: Original Platform
This commit is contained in:
ModelHub XC
2026-09-03 01:40:13 +08:00
commit f9f9aad288
30 changed files with 301 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
Qwen_Qwen2.5-VL-7B-Instruct-IQ2_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:36c02c625faf78240c09b19d48de8da61dceeeb0de09a5c533d4feaea8800b5e
size 2780341056

3
Qwen_Qwen2.5-VL-7B-Instruct-IQ3_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:b49fb8ab4ff54dbda6f46282233c294fd0b5791b0f1b6697974f4d1c2c14e88c
size 3574010688

3
Qwen_Qwen2.5-VL-7B-Instruct-IQ3_XS.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:425eb25ef4caff50b1959b24fe1eb6b76f950c39e77058b0bb465b8eb4e77f13
size 3346254656

3
Qwen_Qwen2.5-VL-7B-Instruct-IQ3_XXS.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:d4a8d6f78e6a4f42bae4968f03611d94c7873d4ce4ccf5f1c1033efe8b760a1c
size 3114513216

3
Qwen_Qwen2.5-VL-7B-Instruct-IQ4_NL.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:8780f9b186905d3ba23b602da3106e097b5734a215aa354f94b5b64c0b73c5cb
size 4437812032

3
Qwen_Qwen2.5-VL-7B-Instruct-IQ4_XS.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:8ac20486d10233a92fc6d1dff00a357d0baca47137a72550db9e87d0a2cd1fae
size 4218471232

3
Qwen_Qwen2.5-VL-7B-Instruct-Q2_K.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:3953245837a33a424e9afc176b3a78d0450f9935f665f217b92448ac6f67dba2
size 3015938880

3
Qwen_Qwen2.5-VL-7B-Instruct-Q2_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:e7a6c87bf8bc353d5b40eb30b2f9389559196747f154d439f4c312c2107879b8
size 3548162880

3
Qwen_Qwen2.5-VL-7B-Instruct-Q3_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:e1ce62ba623b00dd2b0b3563310d04534682df529fd41e25a4cc69203078fe18
size 4088458048

3
Qwen_Qwen2.5-VL-7B-Instruct-Q3_K_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:b45973198cd3867775e994338f19fe19b14a0f26a9e66e6e30582031700059ef
size 3808389952

3
Qwen_Qwen2.5-VL-7B-Instruct-Q3_K_S.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:d8897583d482ecf8d1ddf2db26ccf558f9db9e9d8ebd530a03d7ffa73c43374d
size 3492367168

3
Qwen_Qwen2.5-VL-7B-Instruct-Q3_K_XL.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:d855e27e9c9cc925642942e4b81709e202134a01d1165d6e62fe01854058cbb4
size 4565330752

3
Qwen_Qwen2.5-VL-7B-Instruct-Q4_0.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:ce54f09487edf01bccca242a488012d1764d8813de33d15195b5fb39021df733
size 4444119872

3
Qwen_Qwen2.5-VL-7B-Instruct-Q4_1.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:536ef3c648b6ed52a63349626c28dc1238fd1c4229e5a6db20bd0c62332a5360
size 4873282368

3
Qwen_Qwen2.5-VL-7B-Instruct-Q4_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:c6eaf7638b4b310627467721f528fae1084673d0dc9ee8776c4989c3ac7359af
size 5087562560

3
Qwen_Qwen2.5-VL-7B-Instruct-Q4_K_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:3f4513330aa7f109922bd701d773575484ae2b4a4090d6511260a2a4f8e3d069
size 4683072320

3
Qwen_Qwen2.5-VL-7B-Instruct-Q4_K_S.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:8b99bfd08f9acf248f4875efa1bf9ed164791cdfe3c1a5f6f61a4a1331392493
size 4457767744

3
Qwen_Qwen2.5-VL-7B-Instruct-Q5_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:ff91a101bed09ff78346f7854c030aac322eb14b630c013421237e8d7c431c30
size 5781195584

3
Qwen_Qwen2.5-VL-7B-Instruct-Q5_K_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:325935d89110f25765b1627de643fcb7dbb4a9a52ce3da1820fce170cb0ed4fb
size 5444830016

3
Qwen_Qwen2.5-VL-7B-Instruct-Q5_K_S.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:6c0a4f7edd8536ad56ff4fb326b424e03adddea77a1f43c84bd78574365893c0
size 5315175232

3
Qwen_Qwen2.5-VL-7B-Instruct-Q6_K.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:9cbdf6e5b899f5a5a65d904ea7891284768caeacfb1e5e2a6deb7406ecb71dac
size 6254197568

3
Qwen_Qwen2.5-VL-7B-Instruct-Q6_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:ab8d1f2a34f2042cf6c0e0afd6bfaf5ee81a743b5f7226f4e4f6b5dbecdda8d4
size 6518180672

3
Qwen_Qwen2.5-VL-7B-Instruct-Q8_0.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:c5c575a5b02cbf08773378c35a6d065e4b9b72b2247406bd0134d11f5a549123
size 8098523968

3
Qwen_Qwen2.5-VL-7B-Instruct-bf16.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:4a88f3ed9d639e9e2945b53ec3a595fb0a619d3ca684a62f0764149c7d0b17af
size 15237851680

BIN
Qwen_Qwen2.5-VL-7B-Instruct.imatrix Normal file
View File

Binary file not shown.

175
README.md Normal file
View File

@@ -0,0 +1,175 @@
---
quantized_by: bartowski
pipeline_tag: image-text-to-text
language:
- en
license: apache-2.0
base_model_relation: quantized
base_model: Qwen/Qwen2.5-VL-7B-Instruct
tags:
- multimodal
---
## Llamacpp imatrix Quantizations of Qwen2.5-VL-7B-Instruct by Qwen
Using <a href="https://github.com/ggerganov/llama.cpp/">llama.cpp</a> release <a href="https://github.com/ggerganov/llama.cpp/releases/tag/b5317">b5317</a> for quantization.
Original model: https://huggingface.co/Qwen/Qwen2.5-VL-7B-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 |
| -------- | ---------- | --------- | ----- | ----------- |
| [Qwen2.5-VL-7B-Instruct-bf16.gguf](https://huggingface.co/bartowski/Qwen_Qwen2.5-VL-7B-Instruct-GGUF/blob/main/Qwen_Qwen2.5-VL-7B-Instruct-bf16.gguf) | bf16 | 15.24GB | false | Full BF16 weights. |
| [Qwen2.5-VL-7B-Instruct-Q8_0.gguf](https://huggingface.co/bartowski/Qwen_Qwen2.5-VL-7B-Instruct-GGUF/blob/main/Qwen_Qwen2.5-VL-7B-Instruct-Q8_0.gguf) | Q8_0 | 8.10GB | false | Extremely high quality, generally unneeded but max available quant. |
| [Qwen2.5-VL-7B-Instruct-Q6_K_L.gguf](https://huggingface.co/bartowski/Qwen_Qwen2.5-VL-7B-Instruct-GGUF/blob/main/Qwen_Qwen2.5-VL-7B-Instruct-Q6_K_L.gguf) | Q6_K_L | 6.52GB | false | Uses Q8_0 for embed and output weights. Very high quality, near perfect, *recommended*. |
| [Qwen2.5-VL-7B-Instruct-Q6_K.gguf](https://huggingface.co/bartowski/Qwen_Qwen2.5-VL-7B-Instruct-GGUF/blob/main/Qwen_Qwen2.5-VL-7B-Instruct-Q6_K.gguf) | Q6_K | 6.25GB | false | Very high quality, near perfect, *recommended*. |
| [Qwen2.5-VL-7B-Instruct-Q5_K_L.gguf](https://huggingface.co/bartowski/Qwen_Qwen2.5-VL-7B-Instruct-GGUF/blob/main/Qwen_Qwen2.5-VL-7B-Instruct-Q5_K_L.gguf) | Q5_K_L | 5.78GB | false | Uses Q8_0 for embed and output weights. High quality, *recommended*. |
| [Qwen2.5-VL-7B-Instruct-Q5_K_M.gguf](https://huggingface.co/bartowski/Qwen_Qwen2.5-VL-7B-Instruct-GGUF/blob/main/Qwen_Qwen2.5-VL-7B-Instruct-Q5_K_M.gguf) | Q5_K_M | 5.44GB | false | High quality, *recommended*. |
| [Qwen2.5-VL-7B-Instruct-Q5_K_S.gguf](https://huggingface.co/bartowski/Qwen_Qwen2.5-VL-7B-Instruct-GGUF/blob/main/Qwen_Qwen2.5-VL-7B-Instruct-Q5_K_S.gguf) | Q5_K_S | 5.32GB | false | High quality, *recommended*. |
| [Qwen2.5-VL-7B-Instruct-Q4_K_L.gguf](https://huggingface.co/bartowski/Qwen_Qwen2.5-VL-7B-Instruct-GGUF/blob/main/Qwen_Qwen2.5-VL-7B-Instruct-Q4_K_L.gguf) | Q4_K_L | 5.09GB | false | Uses Q8_0 for embed and output weights. Good quality, *recommended*. |
| [Qwen2.5-VL-7B-Instruct-Q4_1.gguf](https://huggingface.co/bartowski/Qwen_Qwen2.5-VL-7B-Instruct-GGUF/blob/main/Qwen_Qwen2.5-VL-7B-Instruct-Q4_1.gguf) | Q4_1 | 4.87GB | false | Legacy format, similar performance to Q4_K_S but with improved tokens/watt on Apple silicon. |
| [Qwen2.5-VL-7B-Instruct-Q4_K_M.gguf](https://huggingface.co/bartowski/Qwen_Qwen2.5-VL-7B-Instruct-GGUF/blob/main/Qwen_Qwen2.5-VL-7B-Instruct-Q4_K_M.gguf) | Q4_K_M | 4.68GB | false | Good quality, default size for most use cases, *recommended*. |
| [Qwen2.5-VL-7B-Instruct-Q3_K_XL.gguf](https://huggingface.co/bartowski/Qwen_Qwen2.5-VL-7B-Instruct-GGUF/blob/main/Qwen_Qwen2.5-VL-7B-Instruct-Q3_K_XL.gguf) | Q3_K_XL | 4.57GB | false | Uses Q8_0 for embed and output weights. Lower quality but usable, good for low RAM availability. |
| [Qwen2.5-VL-7B-Instruct-Q4_K_S.gguf](https://huggingface.co/bartowski/Qwen_Qwen2.5-VL-7B-Instruct-GGUF/blob/main/Qwen_Qwen2.5-VL-7B-Instruct-Q4_K_S.gguf) | Q4_K_S | 4.46GB | false | Slightly lower quality with more space savings, *recommended*. |
| [Qwen2.5-VL-7B-Instruct-Q4_0.gguf](https://huggingface.co/bartowski/Qwen_Qwen2.5-VL-7B-Instruct-GGUF/blob/main/Qwen_Qwen2.5-VL-7B-Instruct-Q4_0.gguf) | Q4_0 | 4.44GB | false | Legacy format, offers online repacking for ARM and AVX CPU inference. |
| [Qwen2.5-VL-7B-Instruct-IQ4_NL.gguf](https://huggingface.co/bartowski/Qwen_Qwen2.5-VL-7B-Instruct-GGUF/blob/main/Qwen_Qwen2.5-VL-7B-Instruct-IQ4_NL.gguf) | IQ4_NL | 4.44GB | false | Similar to IQ4_XS, but slightly larger. Offers online repacking for ARM CPU inference. |
| [Qwen2.5-VL-7B-Instruct-IQ4_XS.gguf](https://huggingface.co/bartowski/Qwen_Qwen2.5-VL-7B-Instruct-GGUF/blob/main/Qwen_Qwen2.5-VL-7B-Instruct-IQ4_XS.gguf) | IQ4_XS | 4.22GB | false | Decent quality, smaller than Q4_K_S with similar performance, *recommended*. |
| [Qwen2.5-VL-7B-Instruct-Q3_K_L.gguf](https://huggingface.co/bartowski/Qwen_Qwen2.5-VL-7B-Instruct-GGUF/blob/main/Qwen_Qwen2.5-VL-7B-Instruct-Q3_K_L.gguf) | Q3_K_L | 4.09GB | false | Lower quality but usable, good for low RAM availability. |
| [Qwen2.5-VL-7B-Instruct-Q3_K_M.gguf](https://huggingface.co/bartowski/Qwen_Qwen2.5-VL-7B-Instruct-GGUF/blob/main/Qwen_Qwen2.5-VL-7B-Instruct-Q3_K_M.gguf) | Q3_K_M | 3.81GB | false | Low quality. |
| [Qwen2.5-VL-7B-Instruct-IQ3_M.gguf](https://huggingface.co/bartowski/Qwen_Qwen2.5-VL-7B-Instruct-GGUF/blob/main/Qwen_Qwen2.5-VL-7B-Instruct-IQ3_M.gguf) | IQ3_M | 3.57GB | false | Medium-low quality, new method with decent performance comparable to Q3_K_M. |
| [Qwen2.5-VL-7B-Instruct-Q2_K_L.gguf](https://huggingface.co/bartowski/Qwen_Qwen2.5-VL-7B-Instruct-GGUF/blob/main/Qwen_Qwen2.5-VL-7B-Instruct-Q2_K_L.gguf) | Q2_K_L | 3.55GB | false | Uses Q8_0 for embed and output weights. Very low quality but surprisingly usable. |
| [Qwen2.5-VL-7B-Instruct-Q3_K_S.gguf](https://huggingface.co/bartowski/Qwen_Qwen2.5-VL-7B-Instruct-GGUF/blob/main/Qwen_Qwen2.5-VL-7B-Instruct-Q3_K_S.gguf) | Q3_K_S | 3.49GB | false | Low quality, not recommended. |
| [Qwen2.5-VL-7B-Instruct-IQ3_XS.gguf](https://huggingface.co/bartowski/Qwen_Qwen2.5-VL-7B-Instruct-GGUF/blob/main/Qwen_Qwen2.5-VL-7B-Instruct-IQ3_XS.gguf) | IQ3_XS | 3.35GB | false | Lower quality, new method with decent performance, slightly better than Q3_K_S. |
| [Qwen2.5-VL-7B-Instruct-IQ3_XXS.gguf](https://huggingface.co/bartowski/Qwen_Qwen2.5-VL-7B-Instruct-GGUF/blob/main/Qwen_Qwen2.5-VL-7B-Instruct-IQ3_XXS.gguf) | IQ3_XXS | 3.11GB | false | Lower quality, new method with decent performance, comparable to Q3 quants. |
| [Qwen2.5-VL-7B-Instruct-Q2_K.gguf](https://huggingface.co/bartowski/Qwen_Qwen2.5-VL-7B-Instruct-GGUF/blob/main/Qwen_Qwen2.5-VL-7B-Instruct-Q2_K.gguf) | Q2_K | 3.02GB | false | Very low quality but surprisingly usable. |
| [Qwen2.5-VL-7B-Instruct-IQ2_M.gguf](https://huggingface.co/bartowski/Qwen_Qwen2.5-VL-7B-Instruct-GGUF/blob/main/Qwen_Qwen2.5-VL-7B-Instruct-IQ2_M.gguf) | IQ2_M | 2.78GB | 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/Qwen_Qwen2.5-VL-7B-Instruct-GGUF --include "Qwen_Qwen2.5-VL-7B-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/Qwen_Qwen2.5-VL-7B-Instruct-GGUF --include "Qwen_Qwen2.5-VL-7B-Instruct-Q8_0/*" --local-dir ./
```
You can either specify a new local-dir (Qwen_Qwen2.5-VL-7B-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": "image-text-to-text", "allow_remote": true}

3
mmproj-Qwen_Qwen2.5-VL-7B-Instruct-bf16.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:d1c7588c0bdf6e7889737c01cfd54309240d54042e102275880a514ae979aea3
size 1354162912

3
mmproj-Qwen_Qwen2.5-VL-7B-Instruct-f16.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:c24a7f5fcfc68286f0a217023b6738e73bea4f11787a43e8238d4bb1b8604cde
size 1354162912