Method _add_batch_token

examples/server/server.py:1429–1445

(
        self,
        *,
        token: int,
        pos: int,
        seq_id: int,
        logits: bool,
    )

        )

    def _add_batch_token(
        self,
        *,
        token: int,
        pos: int,
        seq_id: int,
        logits: bool,
    ) -> None:
        slot = int(self.batch.n_tokens)
        if slot >= self.n_batch:
            raise RuntimeError("MTP draft batch capacity exceeded")
        self.batch.token[slot] = int(token)
        self.batch.pos[slot] = int(pos)
        self.batch.seq_id[slot][0] = int(seq_id)
        self.batch.n_seq_id[slot] = 1
        self.batch.logits[slot] = int(logits)
        self.batch.n_tokens += 1

    def _try_decode_batch(self) -> bool:
        n_tokens = int(self.batch.n_tokens)