Upload README.md with huggingface_hub

README.md

@@ -1,200 +1,127 @@
 ---
-base_model: Qwen/Qwen2.5-Coder-32B-Instruct
-pipeline_tag: text-generation
 quantized_by: bartowski
+pipeline_tag: text-generation
 ---
 
-# Model Card for Model ID
-
-<!-- Provide a quick summary of what the model is/does. -->
-
-
-
-## Model Details
-
-### Model Description
-
-<!-- Provide a longer summary of what this model is. -->
-
-
-
-- **Developed by:** [More Information Needed]
-- **Funded by [optional]:** [More Information Needed]
-- **Shared by [optional]:** [More Information Needed]
-- **Model type:** [More Information Needed]
-- **Language(s) (NLP):** [More Information Needed]
-- **License:** [More Information Needed]
-- **Finetuned from model [optional]:** [More Information Needed]
-
-### Model Sources [optional]
-
-<!-- Provide the basic links for the model. -->
-
-- **Repository:** [More Information Needed]
-- **Paper [optional]:** [More Information Needed]
-- **Demo [optional]:** [More Information Needed]
-
-## Uses
-
-<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
-
-### Direct Use
-
-<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
-
-[More Information Needed]
-
-### Downstream Use [optional]
-
-<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
-
-[More Information Needed]
-
-### Out-of-Scope Use
-
-<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
-
-[More Information Needed]
-
-## Bias, Risks, and Limitations
-
-<!-- This section is meant to convey both technical and sociotechnical limitations. -->
-
-[More Information Needed]
-
-### Recommendations
-
-<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
-
-Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
-
-## How to Get Started with the Model
-
-Use the code below to get started with the model.
-
-[More Information Needed]
-
-## Training Details
-
-### Training Data
-
-<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
-
-[More Information Needed]
-
-### Training Procedure
-
-<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
-
-#### Preprocessing [optional]
-
-[More Information Needed]
-
-
-#### Training Hyperparameters
-
-- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
-
-#### Speeds, Sizes, Times [optional]
-
-<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
-
-[More Information Needed]
-
-## Evaluation
-
-<!-- This section describes the evaluation protocols and provides the results. -->
-
-### Testing Data, Factors & Metrics
-
-#### Testing Data
-
-<!-- This should link to a Dataset Card if possible. -->
-
-[More Information Needed]
-
-#### Factors
-
-<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+## Llamacpp imatrix Quantizations of Qwen2.5-Coder-32B-Instruct
 
-[More Information Needed]
+Using <a href="https://github.com/ggerganov/llama.cpp/">llama.cpp</a> release <a href="https://github.com/ggerganov/llama.cpp/releases/tag/b4014">b4014</a> for quantization.
 
-#### Metrics
+Original model: https://huggingface.co/Qwen/Qwen2.5-Coder-32B-Instruct
 
-<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+All quants made using imatrix option with dataset from [here](https://gist.github.com/bartowski1182/eb213dccb3571f863da82e99418f81e8)
 
-[More Information Needed]
+Run them in [LM Studio](https://lmstudio.ai/)
 
-### Results
+## Prompt format
 
-[More Information Needed]
+```
+<|im_start|>system
+{system_prompt}<|im_end|>
+<|im_start|>user
+{prompt}<|im_end|>
+<|im_start|>assistant
+```
 
-#### Summary
+## Download a file (not the whole branch) from below:
 
+| Filename | Quant type | File Size | Split | Description |
+| -------- | ---------- | --------- | ----- | ----------- |
+| [Qwen2.5-Coder-32B-Instruct-Q8_0.gguf](https://huggingface.co/bartowski/Qwen2.5-Coder-32B-Instruct-GGUF/blob/main/Qwen2.5-Coder-32B-Instruct-Q8_0.gguf) | Q8_0 | 34.82GB | false | Extremely high quality, generally unneeded but max available quant. |
+| [Qwen2.5-Coder-32B-Instruct-Q6_K_L.gguf](https://huggingface.co/bartowski/Qwen2.5-Coder-32B-Instruct-GGUF/blob/main/Qwen2.5-Coder-32B-Instruct-Q6_K_L.gguf) | Q6_K_L | 27.26GB | false | Uses Q8_0 for embed and output weights. Very high quality, near perfect, *recommended*. |
+| [Qwen2.5-Coder-32B-Instruct-Q6_K.gguf](https://huggingface.co/bartowski/Qwen2.5-Coder-32B-Instruct-GGUF/blob/main/Qwen2.5-Coder-32B-Instruct-Q6_K.gguf) | Q6_K | 26.89GB | false | Very high quality, near perfect, *recommended*. |
+| [Qwen2.5-Coder-32B-Instruct-Q5_K_L.gguf](https://huggingface.co/bartowski/Qwen2.5-Coder-32B-Instruct-GGUF/blob/main/Qwen2.5-Coder-32B-Instruct-Q5_K_L.gguf) | Q5_K_L | 23.74GB | false | Uses Q8_0 for embed and output weights. High quality, *recommended*. |
+| [Qwen2.5-Coder-32B-Instruct-Q5_K_M.gguf](https://huggingface.co/bartowski/Qwen2.5-Coder-32B-Instruct-GGUF/blob/main/Qwen2.5-Coder-32B-Instruct-Q5_K_M.gguf) | Q5_K_M | 23.26GB | false | High quality, *recommended*. |
+| [Qwen2.5-Coder-32B-Instruct-Q5_K_S.gguf](https://huggingface.co/bartowski/Qwen2.5-Coder-32B-Instruct-GGUF/blob/main/Qwen2.5-Coder-32B-Instruct-Q5_K_S.gguf) | Q5_K_S | 22.64GB | false | High quality, *recommended*. |
+| [Qwen2.5-Coder-32B-Instruct-Q4_K_L.gguf](https://huggingface.co/bartowski/Qwen2.5-Coder-32B-Instruct-GGUF/blob/main/Qwen2.5-Coder-32B-Instruct-Q4_K_L.gguf) | Q4_K_L | 20.43GB | false | Uses Q8_0 for embed and output weights. Good quality, *recommended*. |
+| [Qwen2.5-Coder-32B-Instruct-Q4_K_M.gguf](https://huggingface.co/bartowski/Qwen2.5-Coder-32B-Instruct-GGUF/blob/main/Qwen2.5-Coder-32B-Instruct-Q4_K_M.gguf) | Q4_K_M | 19.85GB | false | Good quality, default size for must use cases, *recommended*. |
+| [Qwen2.5-Coder-32B-Instruct-Q4_K_S.gguf](https://huggingface.co/bartowski/Qwen2.5-Coder-32B-Instruct-GGUF/blob/main/Qwen2.5-Coder-32B-Instruct-Q4_K_S.gguf) | Q4_K_S | 18.78GB | false | Slightly lower quality with more space savings, *recommended*. |
+| [Qwen2.5-Coder-32B-Instruct-Q4_0.gguf](https://huggingface.co/bartowski/Qwen2.5-Coder-32B-Instruct-GGUF/blob/main/Qwen2.5-Coder-32B-Instruct-Q4_0.gguf) | Q4_0 | 18.71GB | false | Legacy format, generally not worth using over similarly sized formats |
+| [Qwen2.5-Coder-32B-Instruct-IQ4_NL.gguf](https://huggingface.co/bartowski/Qwen2.5-Coder-32B-Instruct-GGUF/blob/main/Qwen2.5-Coder-32B-Instruct-IQ4_NL.gguf) | IQ4_NL | 18.68GB | false | Similar to IQ4_XS, but slightly larger. |
+| [Qwen2.5-Coder-32B-Instruct-Q4_0_8_8.gguf](https://huggingface.co/bartowski/Qwen2.5-Coder-32B-Instruct-GGUF/blob/main/Qwen2.5-Coder-32B-Instruct-Q4_0_8_8.gguf) | Q4_0_8_8 | 18.64GB | false | Optimized for ARM inference. Requires 'sve' support (see link below). *Don't use on Mac or Windows*. |
+| [Qwen2.5-Coder-32B-Instruct-Q4_0_4_8.gguf](https://huggingface.co/bartowski/Qwen2.5-Coder-32B-Instruct-GGUF/blob/main/Qwen2.5-Coder-32B-Instruct-Q4_0_4_8.gguf) | Q4_0_4_8 | 18.64GB | false | Optimized for ARM inference. Requires 'i8mm' support (see link below). *Don't use on Mac or Windows*. |
+| [Qwen2.5-Coder-32B-Instruct-Q4_0_4_4.gguf](https://huggingface.co/bartowski/Qwen2.5-Coder-32B-Instruct-GGUF/blob/main/Qwen2.5-Coder-32B-Instruct-Q4_0_4_4.gguf) | Q4_0_4_4 | 18.64GB | false | Optimized for ARM inference. Should work well on all ARM chips, pick this if you're unsure. *Don't use on Mac or Windows*. |
+| [Qwen2.5-Coder-32B-Instruct-Q3_K_XL.gguf](https://huggingface.co/bartowski/Qwen2.5-Coder-32B-Instruct-GGUF/blob/main/Qwen2.5-Coder-32B-Instruct-Q3_K_XL.gguf) | Q3_K_XL | 17.93GB | false | Uses Q8_0 for embed and output weights. Lower quality but usable, good for low RAM availability. |
+| [Qwen2.5-Coder-32B-Instruct-IQ4_XS.gguf](https://huggingface.co/bartowski/Qwen2.5-Coder-32B-Instruct-GGUF/blob/main/Qwen2.5-Coder-32B-Instruct-IQ4_XS.gguf) | IQ4_XS | 17.69GB | false | Decent quality, smaller than Q4_K_S with similar performance, *recommended*. |
+| [Qwen2.5-Coder-32B-Instruct-Q3_K_L.gguf](https://huggingface.co/bartowski/Qwen2.5-Coder-32B-Instruct-GGUF/blob/main/Qwen2.5-Coder-32B-Instruct-Q3_K_L.gguf) | Q3_K_L | 17.25GB | false | Lower quality but usable, good for low RAM availability. |
+| [Qwen2.5-Coder-32B-Instruct-Q3_K_M.gguf](https://huggingface.co/bartowski/Qwen2.5-Coder-32B-Instruct-GGUF/blob/main/Qwen2.5-Coder-32B-Instruct-Q3_K_M.gguf) | Q3_K_M | 15.94GB | false | Low quality. |
+| [Qwen2.5-Coder-32B-Instruct-IQ3_M.gguf](https://huggingface.co/bartowski/Qwen2.5-Coder-32B-Instruct-GGUF/blob/main/Qwen2.5-Coder-32B-Instruct-IQ3_M.gguf) | IQ3_M | 14.81GB | false | Medium-low quality, new method with decent performance comparable to Q3_K_M. |
+| [Qwen2.5-Coder-32B-Instruct-Q3_K_S.gguf](https://huggingface.co/bartowski/Qwen2.5-Coder-32B-Instruct-GGUF/blob/main/Qwen2.5-Coder-32B-Instruct-Q3_K_S.gguf) | Q3_K_S | 14.39GB | false | Low quality, not recommended. |
+| [Qwen2.5-Coder-32B-Instruct-IQ3_XS.gguf](https://huggingface.co/bartowski/Qwen2.5-Coder-32B-Instruct-GGUF/blob/main/Qwen2.5-Coder-32B-Instruct-IQ3_XS.gguf) | IQ3_XS | 13.71GB | false | Lower quality, new method with decent performance, slightly better than Q3_K_S. |
+| [Qwen2.5-Coder-32B-Instruct-Q2_K_L.gguf](https://huggingface.co/bartowski/Qwen2.5-Coder-32B-Instruct-GGUF/blob/main/Qwen2.5-Coder-32B-Instruct-Q2_K_L.gguf) | Q2_K_L | 13.07GB | false | Uses Q8_0 for embed and output weights. Very low quality but surprisingly usable. |
+| [Qwen2.5-Coder-32B-Instruct-IQ3_XXS.gguf](https://huggingface.co/bartowski/Qwen2.5-Coder-32B-Instruct-GGUF/blob/main/Qwen2.5-Coder-32B-Instruct-IQ3_XXS.gguf) | IQ3_XXS | 12.84GB | false | Lower quality, new method with decent performance, comparable to Q3 quants. |
+| [Qwen2.5-Coder-32B-Instruct-Q2_K.gguf](https://huggingface.co/bartowski/Qwen2.5-Coder-32B-Instruct-GGUF/blob/main/Qwen2.5-Coder-32B-Instruct-Q2_K.gguf) | Q2_K | 12.31GB | false | Very low quality but surprisingly usable. |
+| [Qwen2.5-Coder-32B-Instruct-IQ2_M.gguf](https://huggingface.co/bartowski/Qwen2.5-Coder-32B-Instruct-GGUF/blob/main/Qwen2.5-Coder-32B-Instruct-IQ2_M.gguf) | IQ2_M | 11.26GB | false | Relatively low quality, uses SOTA techniques to be surprisingly usable. |
+| [Qwen2.5-Coder-32B-Instruct-IQ2_S.gguf](https://huggingface.co/bartowski/Qwen2.5-Coder-32B-Instruct-GGUF/blob/main/Qwen2.5-Coder-32B-Instruct-IQ2_S.gguf) | IQ2_S | 10.39GB | false | Low quality, uses SOTA techniques to be usable. |
+| [Qwen2.5-Coder-32B-Instruct-IQ2_XS.gguf](https://huggingface.co/bartowski/Qwen2.5-Coder-32B-Instruct-GGUF/blob/main/Qwen2.5-Coder-32B-Instruct-IQ2_XS.gguf) | IQ2_XS | 9.96GB | false | Low quality, uses SOTA techniques to be usable. |
+| [Qwen2.5-Coder-32B-Instruct-IQ2_XXS.gguf](https://huggingface.co/bartowski/Qwen2.5-Coder-32B-Instruct-GGUF/blob/main/Qwen2.5-Coder-32B-Instruct-IQ2_XXS.gguf) | IQ2_XXS | 9.03GB | false | Very low quality, uses SOTA techniques to be usable. |
 
+## Embed/output weights
 
-## Model Examination [optional]
+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.
 
-<!-- Relevant interpretability work for the model goes here -->
+Some say that this improves the quality, others don't notice any difference. If you use these models PLEASE COMMENT with your findings. I would like feedback that these are actually used and useful so I don't keep uploading quants no one is using.
 
-[More Information Needed]
+Thanks!
 
-## Environmental Impact
+## Downloading using huggingface-cli
 
-<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+First, make sure you have hugginface-cli installed:
 
-Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+```
+pip install -U "huggingface_hub[cli]"
+```
 
-- **Hardware Type:** [More Information Needed]
-- **Hours used:** [More Information Needed]
-- **Cloud Provider:** [More Information Needed]
-- **Compute Region:** [More Information Needed]
-- **Carbon Emitted:** [More Information Needed]
+Then, you can target the specific file you want:
 
-## Technical Specifications [optional]
+```
+huggingface-cli download bartowski/Qwen2.5-Coder-32B-Instruct-GGUF --include "Qwen2.5-Coder-32B-Instruct-Q4_K_M.gguf" --local-dir ./
+```
 
-### Model Architecture and Objective
+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:
 
-[More Information Needed]
+```
+huggingface-cli download bartowski/Qwen2.5-Coder-32B-Instruct-GGUF --include "Qwen2.5-Coder-32B-Instruct-Q8_0/*" --local-dir ./
+```
 
-### Compute Infrastructure
+You can either specify a new local-dir (Qwen2.5-Coder-32B-Instruct-Q8_0) or download them all in place (./)
 
-[More Information Needed]
+## Q4_0_X_X
 
-#### Hardware
+These are *NOT* for Metal (Apple) offloading, only ARM chips.
 
-[More Information Needed]
+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)
 
-#### Software
+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!).
 
-[More Information Needed]
+## Which file should I choose?
 
-## Citation [optional]
+A great write up with charts showing various performances is provided by Artefact2 [here](https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9)
 
-<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+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.
 
-**BibTeX:**
+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.
 
-[More Information Needed]
+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.
 
-**APA:**
+Next, you'll need to decide if you want to use an 'I-quant' or a 'K-quant'.
 
-[More Information Needed]
+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.
 
-## Glossary [optional]
+If you want to get more into the weeds, you can check out this extremely useful feature chart:
 
-<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+[llama.cpp feature matrix](https://github.com/ggerganov/llama.cpp/wiki/Feature-matrix)
 
-[More Information Needed]
+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.
 
-## More Information [optional]
+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.
 
-[More Information Needed]
+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.
 
-## Model Card Authors [optional]
+## Credits
 
-[More Information Needed]
+Thank you kalomaze and Dampf for assistance in creating the imatrix calibration dataset
 
-## Model Card Contact
+Thank you ZeroWw for the inspiration to experiment with embed/output
 
-[More Information Needed]
+Want to support my work? Visit my ko-fi page here: https://ko-fi.com/bartowski