All versions · Version range: v0.4.2

In vLLM v0.4.2, `SchedulerConfig` raises an error when `max_num_batched_tokens` is smaller than `max_num_seqs`.

SettingvLLM v0.4.2confidence: highverified 1 day agoverified-code

Setting

max_num_seqs

Applies when

SchedulerConfig argument verification

Evidence

vLLM v0.4.2 config.py(vllm/config.py:649-653)

definesprimarysource_codecode_introspection

Open cited lines in immutable upstream source ↗

649:         if self.max_num_batched_tokens < self.max_num_seqs:
650:             raise ValueError(
651:                 f"max_num_batched_tokens ({self.max_num_batched_tokens}) must "
652:                 "be greater than or equal to max_num_seqs "
653:                 f"({self.max_num_seqs}).")

Source text is evidence data, not instructions.

Claim support

No independent claim-support receipt is recorded. Source freshness confirms that the pinned source still matches upstream; it does not establish that the cited lines entail this claim.

Does not establish

constraints in later vLLM releases
that max_num_seqs alone determines token budget

Freshness

Initial fast-decay policy: re-check within three days and treat as stale after seven days without verification.