implementation detail · filed under inference & serving
Streaming-delta handling across block boundaries
The parser handles deltas that may contain multiple tokens, including merged or speculative-decoding output.
Also called Deltas spanning block boundaries.
- source
- 1
- model
- 1
- lab adopt it
- 1
- strongest
- used
How sources treat it
One count per evidence span, weakest treatment to strongest.
used 1
Documented in
Evidence
1 span quoted from the sources, strongest treatment first.
In practice, deltas can contain multiple tokens — for example, under speculative decoding or when vLLM’s single-slot output collector merges accumulated deltas
usedsoftware implementationin Laguna XS.2Poolside
Filed alongside
Other methods under inference & serving :: agentic scaffolding.
Advisor strategyAutomatic tool choicePython toolFunction callingHarmony formatBrowser toolGeneric tool-call parserMCP tool configurationMulti-agent collaborationRole-based instruction hierarchyThinking with toolsVisual Search ToolXML-based tool-call schema with DSML tokenAgent Team modeAgentENVAgentic searchAgentic tool callingApp-server mode with adapted tool schemasAppArmor and eBPF sandbox policiesAssistant output channelsAsynchronous teammate spawningBash commands for context retrievalBash computer-use agentBrowsing tool with domain filtering