Source · Tier B · Documentation

Batch Invariance (vLLM documentation)

vLLM project. 2026. vLLM documentation (GitHub, docs/features/batch_invariance.md).

Originalhttps://github.com/vllm-project/vllm/blob/main/docs/features/batch_invariance.md
VersionMain branch as viewed on 2026-09-25 (GitHub source file; the rendered page at docs.vllm.ai did not return body text to the fetch tool on 2026-09-23). The feature is described as beta, supported on NVIDIA GPUs of compute capability 8.0 or higher and on Intel XPUs with Triton, and tested on dense and mixture-of-experts models.
Accessed2026-09-25

Cited by

Search

Full search page