Update README.md

README.md

@@ -1,3 +1,30 @@
+# CodeVero 7B - 4-bit Quantized
+
+This is a 4-bit quantized version of CodeLLaMA 7B, prepared using `bitsandbytes` and Hugging Face Transformers.  
+Optimized for inference and fine-tuning in low-resource environments.
+
+## Model Details
+- Base: CodeLLaMA-7B
+- Quantization: bitsandbytes 4-bit (bnb_4bit, NF4)
+- Format: Hugging Face (`.safetensors`)
+- Usage: Transformers
+
+## Example Usage
+
+```python
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+model = AutoModelForCausalLM.from_pretrained("your-username/codevero-7b-4bit", device_map="auto")
+tokenizer = AutoTokenizer.from_pretrained("your-username/codevero-7b-4bit")
+
+prompt = "Write a Python function to calculate factorial."
+inputs = tokenizer(prompt, return_tensors="pt").to("cuda")
+
+outputs = model.generate(**inputs, max_new_tokens=100)
+print(tokenizer.decode(outputs[0]))
+
+
+
 ---
 license: mit
 language:
@@ -6,4 +33,4 @@ base_model:
 - codellama/CodeLlama-7b-hf
 tags:
 - co
----
+---