Method decode

examples/server/server.py:12161–12170

(self)

        ]

    def decode(self) -> None:
        batch = self._embedding_batch if self._embedding_batch is not None else self.batch
        if self.embedding and self.has_encoder:
            operation = "llama_encode"
            result = int(llama_cpp.llama_encode(self.ctx, batch))
        else:
            operation = "llama_decode"
            result = int(llama_cpp.llama_decode(self.ctx, batch))
        if result != 0:
            raise RuntimeError(f"{operation} failed with code {result}")

    def process_draft_batch(self) -> None:
        if self.draft_provider is not None: