Prepare folder structure for 3B and 7B models

- Create 3B/, 7B/, 3B-GGUF/, 7B-GGUF/ folders
- Add README files with usage instructions
- Ready for future model uploads

🤖 Generated with Claude Code

Co-Authored-By: Claude <[email protected]>

3B-GGUF/README.md

@@ -0,0 +1,14 @@
+# AnySecret Assistant - 3B GGUF Models
+
+Quantized GGUF versions of the 3B model for use with llama.cpp and Ollama.
+
+Available quantizations:
+- `anysecret-assistant-3B-Q4_K_M.gguf` - 4-bit quantization (smallest)
+- `anysecret-assistant-3B-Q5_K_M.gguf` - 5-bit quantization (recommended)
+- `anysecret-assistant-3B-Q8_0.gguf` - 8-bit quantization (highest quality)
+
+## Usage with Ollama
+```bash
+wget https://huggingface.co/anysecret-io/anysecret-assistant/resolve/main/3B-GGUF/anysecret-assistant-3B-Q5_K_M.gguf
+ollama create anysecret-3b -f Modelfile
+```

3B/README.md

@@ -0,0 +1,12 @@
+# AnySecret Assistant - Llama-3.2-3B Model
+
+Fine-tuned Llama-3.2-3B model for AnySecret configuration assistance.
+
+## Usage
+```python
+from peft import AutoPeftModelForCausalLM
+from transformers import AutoTokenizer
+
+model = AutoPeftModelForCausalLM.from_pretrained("anysecret-io/anysecret-assistant", subfolder="3B")
+tokenizer = AutoTokenizer.from_pretrained("anysecret-io/anysecret-assistant", subfolder="3B")
+```

7B-GGUF/README.md

@@ -0,0 +1,14 @@
+# AnySecret Assistant - 7B GGUF Models
+
+Quantized GGUF versions of the 7B model for use with llama.cpp and Ollama.
+
+Available quantizations:
+- `anysecret-assistant-7B-Q4_K_M.gguf` - 4-bit quantization (smallest)  
+- `anysecret-assistant-7B-Q5_K_M.gguf` - 5-bit quantization (recommended)
+- `anysecret-assistant-7B-Q8_0.gguf` - 8-bit quantization (highest quality)
+
+## Usage with Ollama
+```bash
+wget https://huggingface.co/anysecret-io/anysecret-assistant/resolve/main/7B-GGUF/anysecret-assistant-7B-Q5_K_M.gguf
+ollama create anysecret-7b -f Modelfile
+```

7B/README.md

@@ -0,0 +1,12 @@
+# AnySecret Assistant - CodeLlama-7B Model
+
+Fine-tuned CodeLlama-7B model for AnySecret configuration assistance.
+
+## Usage
+```python
+from peft import AutoPeftModelForCausalLM
+from transformers import AutoTokenizer
+
+model = AutoPeftModelForCausalLM.from_pretrained("anysecret-io/anysecret-assistant", subfolder="7B")
+tokenizer = AutoTokenizer.from_pretrained("anysecret-io/anysecret-assistant", subfolder="7B")
+```