Search papers, labs, and topics across Lattice.
1
0
3
Slash your LLM inference costs by up to 60% without sacrificing accuracy by dynamically routing tasks to smaller models based on confidence estimates.
