2026FlashMLopen source
FreeToken
v0.1.2
FreeToken is an edge-native Mixture-of-Experts (MoE) serving engine designed to run frontier-scale open-weight models on consumer hardware. It optimizes performance by treating heterogeneous edge resources like GPUs, CPUs, and memory as a unified, elastic inference platform.
Bandwidth-adaptive CPU-GPU co-execution (q* policy)
Semantic-aware caching for agentic context edits
Elastic VRAM re-allocation between expert caches and KV memory
pricingOpen Source (Apache License 2.0)
version0.1.2