[KongchangAI]
Concept边端大模型 / 端侧LLM

Edge LLM

A large language model deployed on terminal devices rather than in the cloud. Constrained by memory, compute, and power consumption, it typically has a smaller parameter scale, a limited context window, and no access to external networks.

Timeline (last 90 days)

Oct 19

SRCoT (Self-Reflective Chain-of-Thought) is an architecture that makes the generation process of edge LLMs traceable and auditable through a 'retrieve—reflect—revise' closed-loop process.

Unverified50%
Oct 19

Edge LLMs are constrained by memory, compute and power consumption, typically have smaller parameter counts and limited context windows, and often cannot access external networks.

Unverified50%

All Facts (2)

Source Articles