Unverified50% confidenceFactExact time
Stratum is a local inference tool that keeps commonly used experts in VRAM and places the remaining experts in system memory, letting the CPU and GPU share the computation, and even places large lookup tables on the SSD.
1
Sources
50%
Confidence
Long-term
Relevance
10/31/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedStrata uses a 'doorbell' mechanism: the GPU identifies the 10 experts needed for each layer and writes them to shared memory, the CPU detects the signal and allocates tasks, and the GPU and CPU process the same layer simultaneously77% similarUnverifiedStrata is an inference engine that lets individual developers run relatively large-scale models locally and achieve considerable throughput.74% similarUnverifiedThe plugin uses local inference, with models running directly on the user's own hardware (CPU/GPU) and all data processing completed on the local machine.68% similarUnverifiedBernstein pointed out that agents expand a single question-and-answer into a complete process of planning, retrieval, tool invocation, execution, and checking, simultaneously driving GPU inference load, CPU demand, memory caching, databases, and storage.66% similarUnverifiedColibri adopts an on-demand expert loading approach, loading the corresponding model shards from disk only when needed, a strategy that is especially well-suited to the MoE (Mixture of Experts) architecture.63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1120199API
curl https://kongchang.com/api/v1/knowledge/claims/1120199MCP
get_claim(id=1120199)