Sign in Subscribe

vLLM

Everything tagged vLLM.

Share your story Subscribe

Featured

IBM makes Spyre a first-class PyTorch backend

IBM makes Spyre a first-class PyTorch backend

DP News TeamOct 26, 2026

Google open-sources TorchTPU, bringing native PyTorch to TPUs

Google open-sources TorchTPU, bringing native PyTorch to TPUs

AWS maps two vLLM KV-transfer paths to separate latency bottlenecks

AWS maps two vLLM KV-transfer paths to separate latency bottlenecks

Latest

News
AWS releases MoE inference changes for SageMaker LMI and vLLM
AWS has released a SageMaker LMI v28 routing update and two fused-MoE changes in vLLM, but its H200 performance figures have not been independently…
DP News Team · 2w ago
News
AWS bundles Ray LLM stack for multi-node inference on EKS
AWS’s multi-node Ray LLM Deep Learning Containers package Ray Serve LLM, vLLM, NIXL and EFA for disaggregated prefill and decode on Amazon EKS.
DP News Team · 2w ago
News
Databricks details Proteus harness for shape-specialized GPU kernels
Databricks says Proteus keeps agent-written GPU kernel proposals separate from controlled validation, repeated timing and scoped context, while…
DP News Team · Sep 21
News
Cohere details a megakernel serving engine for North Mini Code
Cohere released an early, model-specific BF16 serving engine that combines a persistent decode megakernel with continuous batching, paged attention…
DP News Team · Sep 10
News
Cohere details dynamic speculative decoding and contributes the method to vLLM
Cohere detailed a hardware-aware dynamic speculative-decoding technique it says runs about 23% faster than fixed-K decoding on its Command A model.
Dmytro Spodarets · Jul 11
News
Baidu's Unlimited-OCR adds vLLM support and tops Hugging Face trending
Baidu's Unlimited-OCR, a 3B open document-parsing model with a flat KV cache, added vLLM support and became the top-trending model on Hugging Face on…
Dmytro Spodarets · Jun 28
Article
10 MLOps Technologies and Trends to Watch in 2026
These 10 MLOps trends are worth watching in 2026 for their potential to improve agent reliability, evaluation, context management, inference…
Dmytro Spodarets · Jan 5