July 2026

800 posts from 102 blogs.

June 2026 August 2026

12 of 800 posts

vLLM BlogJul 29 Optimizing vLLM on Arm CPUsAn overview of Arm CPU enablement and inference performance optimizations in vLLM.vLLMRead

AI and agents

vLLM BlogJul 28 Parallel All the Way Down: Beyond Single-Token Generation with Speculative DecodingSpeculators and vLLM now support P-EAGLE, DFlash, and DSpark, three parallel drafting algorithms that move beyond sequential token generation to deliver faster, simpler, and more scalable speculative decoding for LLM serving.vLLMRead

Reliability and performance

vLLM BlogJul 27 Kimi K3 Is Here: Efficient Day-0 Support on vLLMvLLM delivers day-0 Kimi K3 serving with hybrid KDA prefix caching, DSpark speculative decoding, production-scale disaggregation, and optimized kernels across NVIDIA and AMD GPUs.vLLMRead

Product and company

vLLM BlogJul 23 Announcing vLLM AFD Plugin: Disaggregating Attention and FFN for Flexible MoE ServingWhat vLLM AFD Plugin adds to the vLLM ecosystem: Attention–FFN disaggregation for MoE serving, GPU and Ascend NPU backends, connector-based execution, and graph and ubatching support.vLLMRead

Machine learning

vLLM BlogJul 23 From Day 0 to Production SLAs: Serving GLM-5.2 on 24 NVIDIA B300 GPUs with vLLMHow we took GLM-5.2-NVFP4 from 40 ms to 17 ms mean TPOT on 24 B300 GPUs with vLLM: P/D disaggregation, MTP speculative decoding, Model Runner V2, and the SLA-first trade-offs behind the final configuration.vLLMRead

Reliability and performance

vLLM BlogJul 22 A Preview of Production-Scale Kimi K3 Support on vLLMA preview of production-scale Kimi K3 support in vLLM, including KDA-aware prefix caching, fused kernels, optimized MXFP4 MoE, multimodal integration, and initial NVIDIA and AMD paths.vLLMRead

Product and company

vLLM BlogJul 21 Beyond a Single Model: Building Mixture-of-Models Systems with vLLM Semantic RoutervLLM Semantic Router is expanding from intelligent routing into a system for building, evaluating, and running Mixture-of-Models.vLLMRead

Developer tools

vLLM BlogJul 16 Keeping vLLM Production Quality: A Look Inside CI, Benchmarking, and the Release ProcessHow vLLM maintains production quality with extensive CI across diverse accelerators, nightly performance benchmark and accuracy evaluation, and a two-week release process.vLLMRead

Reliability and performance

vLLM BlogJul 15 TML Inkling on vLLM: Day-0 Support with Optimized PerformancevLLM brings day-0 support to TML Inkling, a 1T-parameter multimodal model, with MTP, long-context serving, parallelism, and up to 380 tokens per second per user on NVIDIA GB200 GPUs.vLLMRead

Reliability and performance

vLLM BlogJul 14 vLLM x TileRT: Specialized Decode for Latency-Critical ServingvLLM prefill paired with TileRT decode through vLLM V1's connector interface: a specialized, latency-optimized decode engine that coexists with native vLLM decode behind one shared serving layer, with zero changes to vLLM.vLLMRead

Machine learning

vLLM BlogJul 13 EAGLE3 Speculative Decoding on AMD Instinct GPUs: Training and Serving with vLLM and AMD QuarkHow AMD Quark trains, quantizes, and serves EAGLE3 speculative-decoding drafts with vLLM on AMD Instinct GPUs, delivering up to 2.00x throughput gains for Kimi-K2.5 and 1.79x for MiniMax-M2.5.vLLMRead

Machine learning

vLLM BlogJul 10 vime + ROCm: End-to-End RL Post-Training on AMD Instinct™ GPUsAnnouncing ROCm support for vime, now running end-to-end on AMD Instinct MI355X GPUs with prebuilt container.vLLMRead