About
Add EAGLE-3 or draft-model speculative decoding to a Jetson vLLM server when TPOT is the bottleneck.
Capabilities
The crawler did not record capability metadata for this resource. Inspect the endpoint
directly to see what it exposes.
Provenance
Discovered Relayed by agntcy
URN authority urn:air:outshift.io:agntcy:jetson-speculative-decoding
Catalog host outshift.io
Anchor check Not anchored
Last crawled seen 5h ago
Tags
streaming servicesdeep learninglarge language modelsnetworkingmodel optimizationmodel fine tuning