bartowski/Dolphin3.0-Qwen2.5-3b-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/Dolphin3.0-Qwen2.5-3b-GGUF
Source: Original Platform
This commit is contained in:
ModelHub XC
2026-07-16 13:50:05 +08:00
commit 26d9b330e8
28 changed files with 322 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

60
.gitattributes vendored Normal file
View File

@@ -0,0 +1,60 @@
*.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
Dolphin3.0-Qwen2.5-3b-f16.gguf filter=lfs diff=lfs merge=lfs -text
Dolphin3.0-Qwen2.5-3b-Q8_0.gguf filter=lfs diff=lfs merge=lfs -text
Dolphin3.0-Qwen2.5-3b-Q6_K_L.gguf filter=lfs diff=lfs merge=lfs -text
Dolphin3.0-Qwen2.5-3b-Q6_K.gguf filter=lfs diff=lfs merge=lfs -text
Dolphin3.0-Qwen2.5-3b-Q5_K_L.gguf filter=lfs diff=lfs merge=lfs -text
Dolphin3.0-Qwen2.5-3b-Q5_K_M.gguf filter=lfs diff=lfs merge=lfs -text
Dolphin3.0-Qwen2.5-3b-Q5_K_S.gguf filter=lfs diff=lfs merge=lfs -text
Dolphin3.0-Qwen2.5-3b-Q4_K_L.gguf filter=lfs diff=lfs merge=lfs -text
Dolphin3.0-Qwen2.5-3b-Q4_K_M.gguf filter=lfs diff=lfs merge=lfs -text
Dolphin3.0-Qwen2.5-3b-Q4_K_S.gguf filter=lfs diff=lfs merge=lfs -text
Dolphin3.0-Qwen2.5-3b-Q4_1.gguf filter=lfs diff=lfs merge=lfs -text
Dolphin3.0-Qwen2.5-3b-Q4_0.gguf filter=lfs diff=lfs merge=lfs -text
Dolphin3.0-Qwen2.5-3b-IQ4_NL.gguf filter=lfs diff=lfs merge=lfs -text
Dolphin3.0-Qwen2.5-3b-IQ4_XS.gguf filter=lfs diff=lfs merge=lfs -text
Dolphin3.0-Qwen2.5-3b-Q3_K_XL.gguf filter=lfs diff=lfs merge=lfs -text
Dolphin3.0-Qwen2.5-3b-Q3_K_L.gguf filter=lfs diff=lfs merge=lfs -text
Dolphin3.0-Qwen2.5-3b-Q3_K_M.gguf filter=lfs diff=lfs merge=lfs -text
Dolphin3.0-Qwen2.5-3b-IQ3_M.gguf filter=lfs diff=lfs merge=lfs -text
Dolphin3.0-Qwen2.5-3b-Q3_K_S.gguf filter=lfs diff=lfs merge=lfs -text
Dolphin3.0-Qwen2.5-3b-IQ3_XS.gguf filter=lfs diff=lfs merge=lfs -text
Dolphin3.0-Qwen2.5-3b-Q2_K_L.gguf filter=lfs diff=lfs merge=lfs -text
Dolphin3.0-Qwen2.5-3b-Q2_K.gguf filter=lfs diff=lfs merge=lfs -text
Dolphin3.0-Qwen2.5-3b-IQ2_M.gguf filter=lfs diff=lfs merge=lfs -text
Dolphin3.0-Qwen2.5-3b-f32.gguf filter=lfs diff=lfs merge=lfs -text
Dolphin3.0-Qwen2.5-3b.imatrix filter=lfs diff=lfs merge=lfs -text

3
Dolphin3.0-Qwen2.5-3b-IQ2_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:489313b653e2e59dffd1f2bb33d00a99c6f9eca76e97859dbbe648e4075fa1eb
size 1140518880

3
Dolphin3.0-Qwen2.5-3b-IQ3_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:0e027da9b27a69dd0e307a90b0044077c92ef26e0549f0aa1b4eb5aa89d39c7a
size 1488898016

3
Dolphin3.0-Qwen2.5-3b-IQ3_XS.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:f4745782c23ff91ebec5ca51eba499ed7c04a74f18cc8994cb4a53ffcb42b1ba
size 1391839200

3
Dolphin3.0-Qwen2.5-3b-IQ4_NL.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:3093c8a913951e5cad5013e0bd44fda44e6d88ad60a4c68af1f369188b1e764c
size 1825212384

3
Dolphin3.0-Qwen2.5-3b-IQ4_XS.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:6a71a45000a10e618364d24e7adce0a0a7475f842f93d25d6d6b4a859b7b2676
size 1739098080

3
Dolphin3.0-Qwen2.5-3b-Q2_K.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:cb46c184487509f1c1fb3d8500aba1ada82ce46ba0759b066930ebcd03dacd2f
size 1274759136

3
Dolphin3.0-Qwen2.5-3b-Q2_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:870cfacf5dd3b62391c70e5d6911b27c1db80dee90f59cc6c1bd3b7756177b36
size 1350119392

3
Dolphin3.0-Qwen2.5-3b-Q3_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:c826e09003a30603d5760ae5645c9339a7783953279a7171ad047cac2ec01b12
size 1707395040

3
Dolphin3.0-Qwen2.5-3b-Q3_K_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:6abe8a033f9ceceeb91262a3aebf5c7f1bc0dbdf0535e8d1c62d860d5dbeaf5b
size 1590478816

3
Dolphin3.0-Qwen2.5-3b-Q3_K_S.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:ee9b071662a5eae2e2963d5ac7098d4944f34dc0a6b078e8c5b1536dc5e5946e
size 1454360544

3
Dolphin3.0-Qwen2.5-3b-Q3_K_XL.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:aec116112d7165cc48546332f34cbad42bf8d19b2194248f3b0aa550d4cd961d
size 1782755296

3
Dolphin3.0-Qwen2.5-3b-Q4_0.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:2043eae883d6b1624d815ee0087aab1cc4bc03f549bd224c2224729a6fc0a7f7
size 1828489184

3
Dolphin3.0-Qwen2.5-3b-Q4_1.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:ab5c43a2ebe233e4a0af3656c0c84ca6b365c4efb22c3c14abf5a2fc7dbaff3d
size 1996261344

3
Dolphin3.0-Qwen2.5-3b-Q4_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:c045652ee7b2274b8f1c579953bf2c79e933485bc384a40f0fe5b6b553747dcc
size 2005266400

3
Dolphin3.0-Qwen2.5-3b-Q4_K_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:0cb1908c5f444e1dc2c5b5619d62ac4957a22ad39cd42f2d0b48e2d8b1c358ab
size 1929906144

3
Dolphin3.0-Qwen2.5-3b-Q4_K_S.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:08e78b9d85dad5d7964f63b01d69c8ce822081aa1cc74cceee69ea5d92be6554
size 1834387424

3
Dolphin3.0-Qwen2.5-3b-Q5_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:5e456a23fe2b0253235e99ccd08d3856a8c7577443226e6db7d8a005fcd9a565
size 2300178400

3
Dolphin3.0-Qwen2.5-3b-Q5_K_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:96076ef91e06e59bf52a45d9541fec72067265b4e2b0d33ba8485c80ce8ee39e
size 2224818144

3
Dolphin3.0-Qwen2.5-3b-Q5_K_S.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:5bba73416f7e8e59a8cb88d7516904cc03e0cbeca34acf3ffa10e5543f8e90db
size 2169669600

3
Dolphin3.0-Qwen2.5-3b-Q6_K.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:52e5a78637a64cd84f7b09dc66ba9a8b02e7f47755ad0d932b34c45d5ea70567
size 2538162144

3
Dolphin3.0-Qwen2.5-3b-Q6_K_L.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:39506626883850f69f38314e116af1fc7c64ee2f6bab26b9c10831521da3eeae
size 2613522400

3
Dolphin3.0-Qwen2.5-3b-Q8_0.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:9121d2cdbf65ef3c22a6c70c6d740af70e87d2381670e42701d55b290e7357eb
size 3285479392

3
Dolphin3.0-Qwen2.5-3b-f16.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:8bd28efb82806b71939dc4e5498e71377d1ddc46a992bb4727060f04f39aca57
size 6178320352

3
Dolphin3.0-Qwen2.5-3b-f32.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:39aa5b74c8d3670babf59730a9f83297ce26cce2ef68437d8272254aafe17602
size 12349714112

3
Dolphin3.0-Qwen2.5-3b.imatrix Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:f3e6badb65b1dad4792f72582ad0911952e83ed2576033f50a3c790f0277fb91
size 3362990

186
README.md Normal file
View File

@@ -0,0 +1,186 @@
---
quantized_by: bartowski
pipeline_tag: text-generation
language:
- en
license_link: https://huggingface.co/Qwen/Qwen2.5-3B/blob/main/LICENSE
license_name: qwen-research
base_model: cognitivecomputations/Dolphin3.0-Qwen2.5-3b
license: other
datasets:
- OpenCoder-LLM/opc-sft-stage1
- OpenCoder-LLM/opc-sft-stage2
- microsoft/orca-agentinstruct-1M-v1
- microsoft/orca-math-word-problems-200k
- NousResearch/hermes-function-calling-v1
- AI-MO/NuminaMath-CoT
- AI-MO/NuminaMath-TIR
- allenai/tulu-3-sft-mixture
- cognitivecomputations/dolphin-coder
- HuggingFaceTB/smoltalk
- cognitivecomputations/samantha-data
- m-a-p/CodeFeedback-Filtered-Instruction
- m-a-p/Code-Feedback
---
## Llamacpp imatrix Quantizations of Dolphin3.0-Qwen2.5-3b
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/cognitivecomputations/Dolphin3.0-Qwen2.5-3b
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 |
| -------- | ---------- | --------- | ----- | ----------- |
| [Dolphin3.0-Qwen2.5-3b-f32.gguf](https://huggingface.co/bartowski/Dolphin3.0-Qwen2.5-3b-GGUF/blob/main/Dolphin3.0-Qwen2.5-3b-f32.gguf) | f32 | 12.35GB | false | Full F32 weights. |
| [Dolphin3.0-Qwen2.5-3b-f16.gguf](https://huggingface.co/bartowski/Dolphin3.0-Qwen2.5-3b-GGUF/blob/main/Dolphin3.0-Qwen2.5-3b-f16.gguf) | f16 | 6.18GB | false | Full F16 weights. |
| [Dolphin3.0-Qwen2.5-3b-Q8_0.gguf](https://huggingface.co/bartowski/Dolphin3.0-Qwen2.5-3b-GGUF/blob/main/Dolphin3.0-Qwen2.5-3b-Q8_0.gguf) | Q8_0 | 3.29GB | false | Extremely high quality, generally unneeded but max available quant. |
| [Dolphin3.0-Qwen2.5-3b-Q6_K_L.gguf](https://huggingface.co/bartowski/Dolphin3.0-Qwen2.5-3b-GGUF/blob/main/Dolphin3.0-Qwen2.5-3b-Q6_K_L.gguf) | Q6_K_L | 2.61GB | false | Uses Q8_0 for embed and output weights. Very high quality, near perfect, *recommended*. |
| [Dolphin3.0-Qwen2.5-3b-Q6_K.gguf](https://huggingface.co/bartowski/Dolphin3.0-Qwen2.5-3b-GGUF/blob/main/Dolphin3.0-Qwen2.5-3b-Q6_K.gguf) | Q6_K | 2.54GB | false | Very high quality, near perfect, *recommended*. |
| [Dolphin3.0-Qwen2.5-3b-Q5_K_L.gguf](https://huggingface.co/bartowski/Dolphin3.0-Qwen2.5-3b-GGUF/blob/main/Dolphin3.0-Qwen2.5-3b-Q5_K_L.gguf) | Q5_K_L | 2.30GB | false | Uses Q8_0 for embed and output weights. High quality, *recommended*. |
| [Dolphin3.0-Qwen2.5-3b-Q5_K_M.gguf](https://huggingface.co/bartowski/Dolphin3.0-Qwen2.5-3b-GGUF/blob/main/Dolphin3.0-Qwen2.5-3b-Q5_K_M.gguf) | Q5_K_M | 2.22GB | false | High quality, *recommended*. |
| [Dolphin3.0-Qwen2.5-3b-Q5_K_S.gguf](https://huggingface.co/bartowski/Dolphin3.0-Qwen2.5-3b-GGUF/blob/main/Dolphin3.0-Qwen2.5-3b-Q5_K_S.gguf) | Q5_K_S | 2.17GB | false | High quality, *recommended*. |
| [Dolphin3.0-Qwen2.5-3b-Q4_K_L.gguf](https://huggingface.co/bartowski/Dolphin3.0-Qwen2.5-3b-GGUF/blob/main/Dolphin3.0-Qwen2.5-3b-Q4_K_L.gguf) | Q4_K_L | 2.01GB | false | Uses Q8_0 for embed and output weights. Good quality, *recommended*. |
| [Dolphin3.0-Qwen2.5-3b-Q4_1.gguf](https://huggingface.co/bartowski/Dolphin3.0-Qwen2.5-3b-GGUF/blob/main/Dolphin3.0-Qwen2.5-3b-Q4_1.gguf) | Q4_1 | 2.00GB | false | Legacy format, similar performance to Q4_K_S but with improved tokens/watt on Apple silicon. |
| [Dolphin3.0-Qwen2.5-3b-Q4_K_M.gguf](https://huggingface.co/bartowski/Dolphin3.0-Qwen2.5-3b-GGUF/blob/main/Dolphin3.0-Qwen2.5-3b-Q4_K_M.gguf) | Q4_K_M | 1.93GB | false | Good quality, default size for most use cases, *recommended*. |
| [Dolphin3.0-Qwen2.5-3b-Q4_K_S.gguf](https://huggingface.co/bartowski/Dolphin3.0-Qwen2.5-3b-GGUF/blob/main/Dolphin3.0-Qwen2.5-3b-Q4_K_S.gguf) | Q4_K_S | 1.83GB | false | Slightly lower quality with more space savings, *recommended*. |
| [Dolphin3.0-Qwen2.5-3b-Q4_0.gguf](https://huggingface.co/bartowski/Dolphin3.0-Qwen2.5-3b-GGUF/blob/main/Dolphin3.0-Qwen2.5-3b-Q4_0.gguf) | Q4_0 | 1.83GB | false | Legacy format, offers online repacking for ARM and AVX CPU inference. |
| [Dolphin3.0-Qwen2.5-3b-IQ4_NL.gguf](https://huggingface.co/bartowski/Dolphin3.0-Qwen2.5-3b-GGUF/blob/main/Dolphin3.0-Qwen2.5-3b-IQ4_NL.gguf) | IQ4_NL | 1.83GB | false | Similar to IQ4_XS, but slightly larger. Offers online repacking for ARM CPU inference. |
| [Dolphin3.0-Qwen2.5-3b-Q3_K_XL.gguf](https://huggingface.co/bartowski/Dolphin3.0-Qwen2.5-3b-GGUF/blob/main/Dolphin3.0-Qwen2.5-3b-Q3_K_XL.gguf) | Q3_K_XL | 1.78GB | false | Uses Q8_0 for embed and output weights. Lower quality but usable, good for low RAM availability. |
| [Dolphin3.0-Qwen2.5-3b-IQ4_XS.gguf](https://huggingface.co/bartowski/Dolphin3.0-Qwen2.5-3b-GGUF/blob/main/Dolphin3.0-Qwen2.5-3b-IQ4_XS.gguf) | IQ4_XS | 1.74GB | false | Decent quality, smaller than Q4_K_S with similar performance, *recommended*. |
| [Dolphin3.0-Qwen2.5-3b-Q3_K_L.gguf](https://huggingface.co/bartowski/Dolphin3.0-Qwen2.5-3b-GGUF/blob/main/Dolphin3.0-Qwen2.5-3b-Q3_K_L.gguf) | Q3_K_L | 1.71GB | false | Lower quality but usable, good for low RAM availability. |
| [Dolphin3.0-Qwen2.5-3b-Q3_K_M.gguf](https://huggingface.co/bartowski/Dolphin3.0-Qwen2.5-3b-GGUF/blob/main/Dolphin3.0-Qwen2.5-3b-Q3_K_M.gguf) | Q3_K_M | 1.59GB | false | Low quality. |
| [Dolphin3.0-Qwen2.5-3b-IQ3_M.gguf](https://huggingface.co/bartowski/Dolphin3.0-Qwen2.5-3b-GGUF/blob/main/Dolphin3.0-Qwen2.5-3b-IQ3_M.gguf) | IQ3_M | 1.49GB | false | Medium-low quality, new method with decent performance comparable to Q3_K_M. |
| [Dolphin3.0-Qwen2.5-3b-Q3_K_S.gguf](https://huggingface.co/bartowski/Dolphin3.0-Qwen2.5-3b-GGUF/blob/main/Dolphin3.0-Qwen2.5-3b-Q3_K_S.gguf) | Q3_K_S | 1.45GB | false | Low quality, not recommended. |
| [Dolphin3.0-Qwen2.5-3b-IQ3_XS.gguf](https://huggingface.co/bartowski/Dolphin3.0-Qwen2.5-3b-GGUF/blob/main/Dolphin3.0-Qwen2.5-3b-IQ3_XS.gguf) | IQ3_XS | 1.39GB | false | Lower quality, new method with decent performance, slightly better than Q3_K_S. |
| [Dolphin3.0-Qwen2.5-3b-Q2_K_L.gguf](https://huggingface.co/bartowski/Dolphin3.0-Qwen2.5-3b-GGUF/blob/main/Dolphin3.0-Qwen2.5-3b-Q2_K_L.gguf) | Q2_K_L | 1.35GB | false | Uses Q8_0 for embed and output weights. Very low quality but surprisingly usable. |
| [Dolphin3.0-Qwen2.5-3b-Q2_K.gguf](https://huggingface.co/bartowski/Dolphin3.0-Qwen2.5-3b-GGUF/blob/main/Dolphin3.0-Qwen2.5-3b-Q2_K.gguf) | Q2_K | 1.27GB | false | Very low quality but surprisingly usable. |
| [Dolphin3.0-Qwen2.5-3b-IQ2_M.gguf](https://huggingface.co/bartowski/Dolphin3.0-Qwen2.5-3b-GGUF/blob/main/Dolphin3.0-Qwen2.5-3b-IQ2_M.gguf) | IQ2_M | 1.14GB | 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/Dolphin3.0-Qwen2.5-3b-GGUF --include "Dolphin3.0-Qwen2.5-3b-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/Dolphin3.0-Qwen2.5-3b-GGUF --include "Dolphin3.0-Qwen2.5-3b-Q8_0/*" --local-dir ./
```
You can either specify a new local-dir (Dolphin3.0-Qwen2.5-3b-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}