llama-cpp-d ~master (2026-03-20T18:06:26.5280947)
Dub
Repo
LlamaContext.decode
llama
ctx
LlamaContext
Decodes a token batch; returns 0 on success.
struct
LlamaContext
@
nogc
nothrow
int
decode
(
llama_batch
batch
)
Meta
Source
See Implementation
llama
ctx
LlamaContext
functions
decode
encode
getEmbeddings
getEmbeddingsIth
getEmbeddingsSeq
getLogits
memoryClear
printPerf
stateGetData
stateGetSize
stateLoadFile
stateSaveFile
stateSeqGetData
stateSeqGetSize
stateSeqSetData
stateSetData
mixins
__anonymous
properties
memory
nCtx
poolingType
static functions
fromModel
Decodes a token batch; returns 0 on success.