All versions · Version range: v0.4.2–v0.23.0

max_num_seqs

Maximum number of sequences to process in a scheduler iteration.

SettingvLLM v0.4.2vLLM v0.23.0verified Sep 29

Claim	Semantic layer	Verified
In vLLM v0.4.2, `max_num_seqs` is exposed through the `SchedulerConfig.__init__` Python parameter `max_num_seqs`.	Interpretation & composition	Sep 29
In vLLM v0.4.2 `SchedulerConfig`, `max_num_seqs` is documented and stored as the maximum number of sequences to process in a single scheduler iteration.	Interpretation & composition	Sep 29
In vLLM v0.4.2, `SchedulerConfig` raises an error when `max_num_batched_tokens` is smaller than `max_num_seqs`.	Interpretation & composition	Sep 29
The latest optimization docs snapshot says that decreasing `max_num_seqs` can reduce concurrent requests in a batch and require less KV cache space when preemptions are frequent.	Interpretation & composition	Aug 20

max_num_seqs is exposed as a SchedulerConfig constructor parameter in the v0.4.2 source snapshot.

vllm.config.scheduler_configpartial_promoted_sliceSchedulerConfig constructor surface in vLLM config.py

vLLM v0.4.2 config.py · vllm/config.py:584-609

max_num_seqs is exposed on SchedulerConfig in the v0.23.0 source snapshot.

vllm.config.scheduler_configcomplete_inventorySchedulerConfig dataclass field surface in vllm/config/scheduler.py

vLLM v0.23.0 config scheduler.py · vllm/config/scheduler.py:63-68