Spaces:
Build error
Build error
|
Download README.md from duckeydevv/Koda-IDS-Frontier: direct link, hf CLI and curl.
- Browser
- Download file 2.3 kB
-
https://huggingface.co/spaces/duckeydevv/Koda-IDS-Frontier/resolve/main/README.md
- Command line
-
hf download hf://spaces/duckeydevv/Koda-IDS-Frontier/README.md
-
curl -L -o README.md https://huggingface.co/spaces/duckeydevv/Koda-IDS-Frontier/resolve/main/README.md
2.3 kB
| title: Koda IDS Frontier | |
| emoji: 🛡️ | |
| colorFrom: blue | |
| colorTo: purple | |
| sdk: gradio | |
| sdk_version: 5.0.1 | |
| app_file: app.py | |
| pinned: false | |
| license: apache-2.0 | |
| hardware: zero-gpu | |
| short_description: Defensive cybersecurity reasoning chat with Koda | |
| tags: | |
| - cybersecurity | |
| - ids | |
| - reasoning | |
| - llama-cpp | |
| - zerogpu | |
| # 🛡️ Koda IDS Frontier | |
| **Lightweight open cybersecurity reasoning model** built on Liquid AI’s LFM2.5-2.6B and specialized for defensive security analysis by NetGoat OSS. | |
| This ZeroGPU Space runs the official **Q4_K_M GGUF** (~1.7 GB) with a clean chat interface. | |
| ## What Koda does | |
| - Interprets security telemetry | |
| - Performs incident triage | |
| - Correlates evidence and evaluates competing explanations | |
| - Explicitly separates **observed / inferred / hypothesized / unknown** | |
| - Recommends proportionate defensive investigation steps | |
| - Avoids treating every anomaly as an attack | |
| ## Model | |
| | Property | Value | | |
| |-----------------|------------------------------------| | |
| | Model | [netgoat-ai/koda-ids-frontier](https://huggingface.co/netgoat-ai/koda-ids-frontier) | | |
| | Base | Liquid AI LFM2.5-2.6B | | |
| | Parameters | ~2.6 B | | |
| | Format used | Q4_K_M GGUF | | |
| | License | Apache-2.0 | | |
| | Context | 8 192 tokens | | |
| ## How to use | |
| 1. Describe a security event, alert, or set of telemetry facts in the chat box. | |
| 2. Optionally tweak temperature / top-p / system prompt in the settings accordion. | |
| 3. Press **Analyze** (or Enter). | |
| Example scenarios are provided in the right column. | |
| > **Important**: Model outputs are research / assistive only. | |
| > Always validate findings against real telemetry and appropriate security tooling. | |
| > The model has no live access to your systems unless you explicitly supply the data. | |
| ## Local reproduction | |
| ```bash | |
| git clone https://huggingface.co/spaces/<your-username>/Koda-IDS-Frontier | |
| cd Koda-IDS-Frontier | |
| pip install -r requirements.txt | |
| python app.py | |
| ``` | |
| ## Credits | |
| - Model: [NetGoat OSS](https://huggingface.co/netgoat-ai) | |
| - Base architecture: [Liquid AI](https://huggingface.co/LiquidAI) | |
| - Runtime: llama.cpp / llama-cpp-python + Hugging Face ZeroGPU | |