UAE Sovereign Cloud
EN ع
Sign in
← Catalogue
Tool Open ○ Public · National
vllm-project /

vLLM

High-throughput, memory-efficient inference and serving engine for LLMs.

Serve
VersionReleasedChangelogSigned
2.3.0 2026-05-18 Accuracy +3 points ✓ 9f2a
2.1.0 2026-03-04 Batch API added ✓ c7b0
1.0.0 2025-12-01 Initial release ✓ 1de4