MCPcopy Create free account
hub / github.com/abetlen/llama-cpp-python / kv_cache_seq_keep

Method kv_cache_seq_keep

llama_cpp/_internals.py:307–309  ·  view source on GitHub ↗
(self, seq_id: int)

Source from the content-addressed store, hash-verified

305 llama_cpp.llama_memory_seq_cp(self.memory, seq_id_src, seq_id_dst, p0, p1)
306
307 def kv_cache_seq_keep(self, seq_id: int):
308 assert self.memory is not None, "Memory is not initialized"
309 llama_cpp.llama_memory_seq_keep(self.memory, seq_id)
310
311 def kv_cache_seq_shift(self, seq_id: int, p0: int, p1: int, shift: int):
312 assert self.memory is not None, "Memory is not initialized"

Callers

nothing calls this directly

Calls

no outgoing calls

Tested by

no test coverage detected