Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* 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
+llama32_datafusion.gguf filter=lfs diff=lfs merge=lfs -text

Modelfile

@@ -0,0 +1,28 @@
+FROM llama32_datafusion.gguf
+
+# System prompt for all sessions
+SYSTEM """You are a helpful, concise, and accurate coding assistant specialized in Rust and the DataFusion SQL engine. Always provide high-level, idiomatic Rust code, DataFusion SQL examples, clear documentation, and robust test cases. Your answers should be precise, actionable, and end with '### End'."""
+
+# Prompt template (optional, but recommended for instruct models)
+TEMPLATE """### Instruction:
+{{ .Prompt }}
+
+### Response:
+"""
+
+# Stop sequences to end generation
+PARAMETER stop "### Instruction:"
+PARAMETER stop "### Response:"
+PARAMETER stop "### End"
+
+# Generation parameters to prevent infinite loops
+PARAMETER num_predict 1024
+PARAMETER repeat_penalty 1.2
+PARAMETER temperature 0.7
+PARAMETER top_p 0.9
+
+# Metadata for public sharing (for reference only)
+# TAGS ["llama3", "datafusion", "qa", "rust", "sql", "public"]
+# DESCRIPTION "A fine-tuned LLM specialized in Rust and DataFusion (SQL engine) Q&A. Produces idiomatic Rust code, DataFusion SQL examples, clear documentation, and robust test cases, with robust stop sequences and infinite loop prevention."
+
+LICENSE "llama-3.2"

README.md

@@ -0,0 +1,82 @@
+---
+license: llama3.2
+base_model: yarenty/llama32-datafusion-instruct
+tags:
+- text-generation
+- instruction
+- datafusion
+- rust
+- code
+- gguf
+---
+
+# Llama 3.2 DataFusion Instruct (GGUF)
+
+This repository contains the GGUF version of the `yarenty/llama32-datafusion-instruct` model, quantized for efficient inference on CPU and other compatible hardware.
+
+For full details on the model, including its training procedure, data, intended use, and limitations, please see the **[full model card](https://huggingface.co/yarenty/llama32-datafusion-instruct)**.
+
+## Model Details
+
+- **Base model:** [yarenty/llama32-datafusion-instruct](https://huggingface.co/yarenty/llama32-datafusion-instruct)
+- **Format:** GGUF
+- **Quantization:** `Q4_K_M` (Please verify and change if different)
+
+## Prompt Template
+
+This model follows the same instruction prompt template as the base model:
+
+```
+### Instruction:
+{Your question or instruction here}
+
+### Response:
+```
+
+## Usage
+
+These files are compatible with tools like `llama.cpp` and `Ollama`.
+
+### With Ollama
+
+1.  Create the `Modelfile`:
+    ```
+    FROM ./llama32_datafusion.gguf
+    TEMPLATE """### Instruction:
+    {{ .Prompt }}
+
+    ### Response:
+    """
+    PARAMETER stop "### Instruction:"
+    PARAMETER stop "### Response:"
+    PARAMETER stop "### End"
+    ```
+
+2.  Create and run the Ollama model:
+    ```bash
+    ollama create llama32-datafusion-instruct-gguf -f Modelfile
+    ollama run llama32-datafusion-instruct-gguf "How do I use the Ballista scheduler?"
+    ```
+
+### With llama.cpp
+
+```bash
+./main -m llama32_datafusion.gguf --color -p "### Instruction:\nHow do I use the Ballista scheduler?\n\n### Response:" -n 256 --stop "### Instruction:" --stop "### Response:" --stop "### End"
+```
+
+## Citation
+
+If you use this model, please cite the original base model:
+```
+@misc{yarenty_2025_llama32_datafusion_instruct,
+  author = {yarenty},
+  title = {Llama 3.2 DataFusion Instruct},
+  year = {2025},
+  publisher = {Hugging Face},
+  journal = {Hugging Face repository},
+  howpublished = {\url{https://huggingface.co/yarenty/llama32-datafusion-instruct}}
+}
+```
+
+## Contact
+For questions or feedback, please open an issue on the Hugging Face repository or the [source GitHub repository](https://github.com/yarenty/trainer).

llama32_datafusion.gguf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a6b51b3ff0bf0b555616f515b9f85a8a2da34d11f6fcdc9afd56c9e18bf8603c
+size 2019377280