Add Support for Ollama and OpenAI LLM Providers

.gitignore

@@ -8,4 +8,44 @@ __pycache__
 data/\ndata/\n*.pdf
 .venv-ci/
 data/
-example_output
+example_output
+.vscode
+
+# Generated and output files
+data/generated/
+data/inference/
+data/outputs/
+*.log
+logs/
+.cache/
+__pycache__/
+*.cache
+*.tmp
+*.temp
+inference_outputs/
+model_outputs/
+results/
+outputs/
+
+# Jupyter notebook checkpoints
+.ipynb_checkpoints/
+
+# IDE files
+.vscode/
+.idea/
+*.swp
+*.swo
+
+# OS files
+.DS_Store
+Thumbs.db
+
+{new_directory}
+.env
+# Common misspellings / alternate names for generated artifacts
+inferenced/
+outptu/
+data/inferenced/
+data/outptu/
+inferenced
+outptu

DOCS.md

@@ -417,7 +417,8 @@ synthetic-data-kit create [OPTIONS] INPUT
 
 | Option | Description |
 |--------|-------------|
-| `--type TEXT` | Content type to generate [qa\|summary\|cot] |
+| `--type TEXT` | Content type to generate [qa\|summary\|cot\|cot-enhance\|multimodal-qa] |
+| `-d, --difficulty TEXT` | Question difficulty [easy\|medium\|advanced] (for qa, cot, multimodal-qa) |
 | `-o, --output-dir PATH` | Directory to save generated content |
 | `--api-base TEXT` | VLLM API base URL |
 | `-m, --model TEXT` | Model to use |
@@ -439,6 +440,11 @@ synthetic-data-kit create data/output/document.txt --type summary
 # Generate Chain of Thought (CoT) reasoning examples
 synthetic-data-kit create data/output/document.txt --type cot
 
+# Control difficulty
+synthetic-data-kit create data/output/document.txt --type qa --difficulty medium
+synthetic-data-kit create data/output/document.txt --type cot --difficulty advanced
+synthetic-data-kit create data/output/document.lance --type multimodal-qa --difficulty easy
+
 # Use custom model
 synthetic-data-kit create data/output/document.txt -m "meta-llama/Llama-3.3-8B-Instruct"
 ```