Open-source AI model research and inference engine focused on compact models, aggressive quantization, and efficient local deployment.
AI-Atlas score
| Machine Learning | 67 |
|---|---|
| Computer Vision | 44 |
| Data Science | 22 |
| NLP & LLMs | 70 |
| AI Agents | 58 |
| MLOps | 39 |
Each node is one domain, scored only from evidence in these public repositories. Select one to see why.
Deep, repeated, well-engineered work in this domain.
4 public repositories → 2 with AI work, evaluated. The other 2 (repositories without AI code, forks and empty repositories) are not scored and have no effect on the result. How repositories are selected
Ordered by how much each project adds to the Atlas score, not by its own score. A project with evidence in several domains can rank above one with a slightly higher project score. Choose a domain to focus on the projects with meaningful evidence there.
76% of Atlas Score
MARK is a privacy-first personal AI assistant that integrates local vision-language inference, deterministic knowledge retrieval, voice and vision processing, sandboxed tool execution, a native Model Context Protocol bridge in Rust, and a resilient private transport mesh for multi-device interaction.
24% of Atlas Score
Mark-AI is a repository implementing Sparse Transformer architectures (Multi-Head Latent Attention and Mixture-of-Experts) with quantization to GGUF format for efficient local inference on consumer GPUs, including training scripts, export tools, and a retrieval-augmented generation component.
