Nate@nate_512

フォロー

meshLLM GitHub project

Running a 70B+ model without an 80GB GPU is possible. Mesh LLM distributes inference across the devices you already own. The architecture is clever: if a node can handle the model, it runs locally; if not, it routes to a peer that can; if the model is too large for a single node, it splits the workload.

1mo

Nate@nate_512

フォロー

meshLLM GitHub project

Running a 70B+ model without an 80GB GPU is possible. Mesh LLM distributes inference across the devices you already own. The architecture is clever: if a node can handle the model, it runs locally; if not, it routes to a peer that can; if the model is too large for a single node, it splits the workload.

1mo