Open-Source
Benchmarks, setup guides and working applications from our own research into running AI models on local and edge hardware. Every project is published on GitHub with its code, configuration and documentation.
Local AI Models
Bug Hunt Benchmark
A browser application that downloads models from the Hugging Face Hub, serves them with local runtimes, and scores how well each model finds and fixes deliberate defects in small projects.
Local AI Models
Qwen3.8 Claude Code
Qwen3.8-27B served locally with llama.cpp at 4-bit and 1-bit, benchmarked with and without MTP speculative decoding, and connected to Claude Code through LiteLLM.
Edge AI
Arduino UNO Q Customer Service Agent
Two Arduino UNO Q boards each run Gemma 3 1B. One plays a clothing shop customer and the other the customer service agent, with no cloud service or API key.
Local AI Models
Muse-Glimmer-30B Claude Code
Muse-Glimmer-30B served locally with llama.cpp on an M4 Max, benchmarked with and without dflash speculative decoding, and connected to Claude Code through LiteLLM.
Local AI Models
MoE on Arc
A benchmark harness that measures accuracy and throughput together, with a verified configuration for Qwen3.6-35B-A3B on an Intel Arc 140V integrated GPU through OpenVINO Model Server.
Let us build your future
Bring us a plan or just an idea. We will arrange a discovery meeting with you to go through it.