Roadmap¶
Current status and planned features.
Current Status¶
ModelDock is in active pre-release development (v0.1.x).
| Component | Status |
|---|---|
| Core architecture | Complete |
| Ollama runtime | Shipped |
| Dynamic catalog | Shipped |
| Python SDK | Shipped |
| CLI | Shipped |
| Smart caching | Shipped |
Planned: Runtime Adapters¶
| Runtime | Status |
|---|---|
| Ollama | :material-check-circle:{ style="color: #4caf50" } Shipped |
| LM Studio | :material-clock-outline:{ style="color: #ff9800" } Planned |
| llama.cpp | :material-clock-outline:{ style="color: #ff9800" } Planned |
| Jan AI | :material-clock-outline:{ style="color: #ff9800" } Planned |
| GPT4All | :material-clock-outline:{ style="color: #ff9800" } Planned |
| vLLM | :material-clock-outline:{ style="color: #ff9800" } Planned |
Planned: Features¶
- Automatic model downloads (default
auto_installon) - Searchable model registry from Python
- Bulk installation by category
- Model recommendations for beginners
- Download progress tracking (rich, tqdm, silent)
- Version management and updates
- Smart aliases (
load("llama3")) - Auto model selection based on task
- Offline cache management
- CLI completions
- Plugin hooks (EventPort lifecycle)
Planned: Infrastructure¶
- Cross-platform CI (Windows/macOS/Linux × Python 3.9–3.12)
- Pre-commit hooks (ruff, mypy, bandit)
- Coverage reporting (≥80% target)
- PyPI publishing via trusted publishing
- Documentation site (MkDocs Material)
Versioning¶
Following Semantic Versioning:
- MAJOR — Incompatible API changes
- MINOR — Backward-compatible new functionality
- PATCH — Backward-compatible bug fixes