WRSWorld Research StateContribute
WORLD RESEARCH STATE

Petals: pipeline inference and adaptation across contributed servers

Contribution grade

Collaborative language-model systems · Human-designed research method or resource

Reference score80,000Provisional editorial estimate of this exact contribution; its grade is assessed separately.

What changed

Split pretrained model blocks across volunteer server chains, with client-managed embeddings, session caches, routing/failover and parameter-efficient adaptation interfaces.

Scientific scope

Split pretrained model blocks across volunteer server chains, with client-managed embeddings, session caches, routing/failover and parameter-efficient adaptation interfaces. Assumes parties trusted with client data and inherits pipeline parallelism/quantization/adapters; not cryptographic privacy or Byzantine correctness. Performance depends on network, server availability and workload.

Evidence

The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. Assumes parties trusted with client data and inherits pipeline parallelism/quantization/adapters; not cryptographic privacy or Byzantine correctness. Performance depends on network, server availability and workload.

Contributors

Provisional WRS assessment, edition 2026-10-16.1. Independent expert calibration is pending. This grade does not certify correctness or assign ownership.

Full scientific record · Browse contribution grades · Request a correction