Add model card

README.md

@@ -0,0 +1,82 @@
+---
+language: en
+license: apache-2.0
+tags:
+  - linux
+  - command-generation
+  - gguf
+  - qwen3
+  - llama-cpp
+  - offline
+base_model: Qwen/Qwen3.5-0.8B
+---
+
+# INCEPT/Sh — Command Inference Model (Q8_0)
+
+Fine-tuned **Qwen3.5-0.8B** (GGUF Q8_0, 774MB) for Linux shell command generation. Maps plain English descriptions to accurate Linux commands. Runs entirely offline via [llama.cpp](https://github.com/ggerganov/llama.cpp).
+
+**Benchmark:** 99/100 on a structured 100-question Linux command evaluation (Ubuntu 22.04, bash, non-root).
+
+## Usage
+
+Requires `llama-server` from [llama.cpp](https://github.com/ggerganov/llama.cpp).
+
+```bash
+# Download model
+huggingface-cli download 0Time/INCEPT-SH \
+  incept-command-v2-q8_0.gguf --local-dir ./models
+
+# Run with INCEPT/Sh CLI
+git clone https://github.com/ProMohanad/INCEPT.sh
+cd INCEPT.sh
+pip install -e ".[cli]"
+incept
+```
+
+Or use directly with llama-server:
+
+```bash
+llama-server -m models/incept-command-v2-q8_0.gguf --port 8787
+```
+
+## Prompt Format
+
+The model uses ChatML format with a system context line:
+
+```
+<|im_start|>system
+ubuntu 22.04 bash non-root
+<|im_end|>
+<|im_start|>user
+{natural language query}
+<|im_end|>
+<|im_start|>assistant
+<think>
+</think>
+
+```
+
+Inference temperature: **0.0** (greedy decoding).
+
+## Training
+
+| Parameter             | Value                                       |
+|-----------------------|---------------------------------------------|
+| Base model            | Qwen/Qwen3.5-0.8B                           |
+| Training method       | Supervised fine-tuning (LoRA, rank 16)      |
+| Training examples     | 79,264 (SFT) + 11,306 (pipe refinement)    |
+| Learning rate         | 5×10⁻⁵                                     |
+| Quantization          | Q8_0 (774MB)                                |
+| Supported distros     | Ubuntu, Debian, RHEL, Arch, Fedora, CentOS  |
+| Training hardware     | Apple M4 Mac mini, 32GB unified RAM         |
+
+## Safety
+
+The accompanying [INCEPT/Sh](https://github.com/ProMohanad/INCEPT.sh) engine includes:
+- Prompt injection detection
+- Catastrophic pattern blocking (fork bombs, `rm -rf /`, pipe-to-shell, etc.)
+- Risk classification: `SAFE` / `CAUTION` / `DANGEROUS` / `BLOCKED`
+
+## License
+
+[Apache 2.0](https://www.apache.org/licenses/LICENSE-2.0)