QVAC Blog

16 June 2026 — We built an app that predicts football matches. The twist: the AI runs entirely on your own machine, it simulates each match 10,000 times, and it reasons out loud before committing to a scoreline. No cloud model, no API key, no per-token bill. You can watch it think. This is a write-up of how it […]

2 June 2026 — TurboQuant is a KV-cache quantization algorithm published by Google Research at ICLR 2026 (Zandieh et al.). It compresses the running context memory of a transformer LLM by up to 5x with nearly no accuracy loss across long-context benchmarks. QVAC SDK 0.12.0 integrates TurboQuant inside qvac-fabric-llm.cpp with a Vulkan backend.

1 June 2026 — QVAC SDK 0.12.0, shipping June 1 2026, integrates TurboQuant, a KV-cache quantization algorithm published by Google Research at ICLR 2026. The result: on-device LLMs can now hold up to 5x more context with the same model, on the same device, with nearly no measurable accuracy loss. No code changes. No retraining.

9 April 2026 — An exploration of the QVAC ecosystem and an invitation to developers who want more than just access to AI, those who want ownership, portability, resilience and the ability to build without asking permission.