vllm serving engine with MLOps experiments

MLOps Tasks & Experiments:

1. Inference Engine - Async/Sync Generator 👨🏻‍💻
2. Embeddings Engine - Sparse, Dense, etc.. 👨🏻‍💻
3. Retriever Engine - Hybrid, etc..👨🏻‍💻
4. HyperParameter fine-tuner [axolotl, unsloth, etc..]
5. Evaluations (seft, peft, etc..)[TODO]

MLOps Providers:

1. BentoML 👨🏻‍💻
2. Ray 👨🏻‍💻
3. ...

Quantization techniques:

1. GPTQ 👨🏻‍💻
2. AWQ 👨🏻‍💻
3. GGUF (locally, smaller devices, etc..) []

@TODO ...