The open-source inference engine vLLM has shipped v0.28.0 (August 26, 2026), and with 584 commits from 270 contributors - 76 of them first-time - the release has one dominant theme: making Kimi-K3 cheaper to run. As one of the most widely deployed open LLM serving engines, vLLM's priorities read like a live snapshot of what self-hosting teams actually need.

The Kimi-K3 full-stack push

The release notes dedicate their longest section to Kimi-K3, and it is a cross-layer effort:

  • Decode Context Parallel (DCP) spreads decode-time long-context load across GPUs;
  • Fused FlashKDA decode/prefill kernels, SiTU activation support for MegaMoE, and GEMM-RS for sequence parallelism;
  • Combined all-gathers with 1.5-3x kernel-level speedup;
  • An adaptive speculative token budget delivering ~60% better DSpark TTFT;
  • Optional shared-expert sharding saving ~17 GiB of memory per GPU;
  • Kimi-K3 now also runs on ROCm via the V2 model runner.

These numbers come from the official release notes - engineering figures, not marketing. Seventeen GiB per GPU means more replicas or longer contexts on the same hardware, and the TTFT improvement maps directly to how long end users wait.

DeepSeek V4 and everything else

On the DeepSeek V4 side, sparse MLA now works end-to-end for plain decode, MTP, and DSpark speculative decoding, joined by AMD Quark NVFP4 support and ROCm enablement on gfx11 and gfx950. Infrastructure-wise: tiered KV cache gained disk offloading, E/P/D disaggregation landed in Model Runner V2, and the default max_num_batched_tokens rose from 8192 to 16384. Newly supported models include Muse Glimmer, Ling 3.0 Flash (with FP8 and hybrid MXFP4 expert variants), Dots3 NOTE, and Interns2mobius. Breaking changes to watch: bitsandbytes moved to an out-of-tree plugin, and Transformers bumped to 5.15.0.

Why it matters

vLLM is the practical default for most self-hosting teams, so its optimization priorities are a ranking of real deployment demand. That Kimi K3 and DeepSeek V4 - two open-weight models from Chinese teams - take the top two spots in a single release says they are past the announcement stage and into large-scale serving. The open-source model race is shifting from launch events to cost per token. (Full details in the GitHub release notes)

So if you self-host open models, this version is worth upgrading to directly; and if you are gauging how deeply Chinese open-weight models have landed, vLLM's commit log is more honest than any launch event.