Kolosal AI
Open-source platform for running large language models locally, so your data stays private and inference stays fast. Custom training and production-ready inference at scale.
Open-source tools, platforms, and applications
Open-source platform for running large language models locally, so your data stays private and inference stays fast. Custom training and production-ready inference at scale.
A Rust/actix-web multimodal inference server: TTS, STT, image classification/detection, and an LLM chat proxy behind adaptive batching, model pooling, and circuit breakers. Built under Kolosal AI.
A ~60k-line Rust AutoML framework: 17+ training algorithms, Bayesian hyperparameter search, drift/anomaly detection, and an ISO-compliance layer, served over an Axum web dashboard and CLI.
A Rust self-organizing map library: 11 initialization strategies, parameter-free density and auto-k routing extensions, and Rayon/Metal/CUDA-accelerated training with built-in cluster quality metrics.
A high-performance nuclear chain reaction simulator written in Rust: GPU-accelerated particle physics, real-time 3D visualization, and a 150+ isotope library drawn from ENDF/B-VIII.0 nuclear data.
A desktop app for evaluating stocks through technical and fundamental analysis, powered by on-device ONNX inference and a local Qwen2.5-0.5B LLM. No cloud, no subscriptions, fully private.
A community-driven platform connecting learners and researchers through structured courses and peer mentorship, built on the belief that good education should be accessible, not gatekept.
Building mental health awareness across Indonesia through accessible psychology content and peer-to-peer support networks, because understanding your mind shouldn't require a degree or a therapist's fee.
An interactive 3D brain visualization tool that predicts cortical activation patterns across six regions in response to text, image, and audio stimuli. It runs LLaMA, CLIP, and Wav2Vec encoders locally in Rust.
A high-performance Rust pipeline for limit order book microstructure research. From a single binary, it ingests raw tick data, computes order-flow imbalance features and 87 technical indicators, runs synthetic LOB simulations, and backtests signal-based strategies.
An authorized web pentest toolkit in Rust: one static binary covers black-box DAST across the OWASP Top 10, local source-code SAST, real-CVE matching, and out-of-band collaboration, detection-only and rate-limited by default.