vLLM-Omni v0.32.0
vLLM-Omni v0.32.0 adds Ray-based multi-node inference, session-aware replica placement, and multi-frontend full-duplex serving.
The Deploy Log | Companies | Free, no signup
Every row from vLLM in The Deploy Log: 6 rows, newest first, each linking the publisher's own page. The newest shipped are vLLM-Omni v0.32.0 and vLLM v0.31.1rc0. 4 of them have a shipped date, the first and the newest . By lane: Tools and products 6. They fall in 2 months. The tables below count them by lane and month.
| Lane | Rows |
|---|---|
| Tools and products | 6 |
| Figure | Value |
|---|---|
| First row dated | |
| Newest row dated | |
| Median days between release days, publisher's dates only | 13 |
| Rows with no publisher's date | 2 |
vLLM-Omni v0.32.0 adds Ray-based multi-node inference, session-aware replica placement, and multi-frontend full-duplex serving.
vLLM v0.31.1rc0 exposes cached prompt tokens by cache tier in metrics.
vLLM v0.31.0 adds a Transformers version upper bound in requirements.
vLLM v0.30.1rc0 adds MI355 dense NVFP4 and MoRI kernel mirrors.
vLLM v0.30.0 fixes DeepGEMM CUDA 12.9 release builds.
vLLM v0.30.0rc2 tagged a bugfix to avoid receive reports for notification-only requests in NIXL.
One line every time AI actually ships, around the clock, free on Threads and Bluesky. Every one is in The Deploy Log with its source.