MATRIX · vLLM

vLLM 0.33 moves to CUDA 13 and adds EFA serving and new inference paths

Mon, 02 Nov 2026 16:56:00 GMT

vLLM 0.33 moves to CUDA 13 and adds EFA serving and new inference paths

Original story: vLLM 0.33 moves stable builds to CUDA 13 and adds EFA serving

Why read: CUDA baselines and serving behavior change together, directly affecting self-hosted upgrades and rollback.

Recommendation: ADOPT

Sources: vLLM 0.33.0 release

vLLM 0.32 tightens security defaults and expands low-precision inference

Sun, 18 Oct 2026 15:58:14 GMT

vLLM 0.32 tightens security defaults and expands low-precision inference

Original story: vLLM 0.33 moves stable builds to CUDA 13 and adds EFA serving

Why read: Security defaults, Python support and streaming contracts change together, requiring a planned production upgrade.

Recommendation: ADOPT

Sources: vLLM 0.32.0 release

vLLM 0.31 extends fast restart and tightens multimodal request boundaries by default

Mon, 05 Oct 2026 16:01:12 GMT

vLLM 0.31 extends fast restart and tightens multimodal request boundaries by default.

Original story: vLLM 0.33 moves stable builds to CUDA 13 and adds EFA serving

Why read: vLLM 0.31 advances both fast recovery and safer multimodal request boundaries.

Recommendation: ADOPT

Sources: vLLM 0.31.0

vLLM 0.33 moves stable builds to CUDA 13 and adds EFA serving

Tue, 22 Sep 2026 16:02:37 GMT

vLLM 0.33 moves its stable CUDA artifacts to CUDA 13, adds a dedicated AWS EFA image for multi-node KV transfer, and makes ROCm 10 the default while retaining ROCm 7.2 builds. It expands speculative decoding, low-precision KV paths, watchdog diagnostics and large-scale serving. The release also removes CUDA 12.9 artifacts and several legacy executors, so deployment compatibility must be tested before rollout.

Recommendation: ADOPT

Sources: vllm-project/vllm · releases / tag / v0.30.0

vLLM 0.28.0 lands a big-model serving push for Kimi-K3 and DeepSeek V4

Sun, 30 Aug 2026 00:00:20 GMT

vLLM 0.28.0 (584 commits from 270 contributors, 76 new) is dominated by a Kimi-K3 performance push: Decode Context Parallel support, fused FlashKDA decode/prefill kernels, combined all-gathers with 1.5-3x kernel-level speedup, an adaptive speculative token budget delivering ~60% better DSpark TTFT, and optional shared-expert sharding saving ~17 GiB per GPU. DeepSeek V4 gets sparse MLA working end-to-end for plain decode, MTP and DSpark speculative decoding, plus AMD Quark NVFP4 and ROCm enablement. Other headline items: DFlash2 speculative decoding with a candidate selector, Model Runner V2 maturation (E/P/D disaggregation, weight offloading), tiered KV-cache offloading including disk, a Rust frontend with gRPC (protobuf schemas on Buf), and new defaults (max_num_batched_tokens 8192->16384, Mamba prefix caching on by default). Breaking changes: bitsandbytes support moved to an out-of-tree plugin, Transformers bumped to 5.15.0, deprecated APIs removed.

Recommendation: ADOPT

Sources: vllm-project/vllm · releases / tag / v0.28.0