#gpu
- AMD and Anthropic announce 2-gigawatt Instinct MI450 compute deal with up to $5B AMD investment Anthropic industry
- SGLang v0.5.11: Speculative Decoding V2 as Default and Eight New Model Architectures tools
- vLLM v0.21.0: Blackwell MLA Backend, HMA KV Offload, Spec Decode for Reasoning Models vLLM Project tools
- vLLM v0.22.0: DeepSeek V4 Production Hardening, Rust Frontend, 28.9% Latency Drop tools
- CUDA-for-AMD-on-Windows stack runs CUDA apps on Radeon via ZLUDA tools
- vLLM v0.20.1: DeepSeek V4 Stabilization on CUDA 13 and PyTorch 2.11 tools
- llama.cpp b9967–b9969: Adreno GPU Acceleration and OpenAI-Compatible Null Sampling tools
- vLLM 0.32.0 widens low-precision KV cache and drops Python 3.10 vLLM tools
- vLLM serves 1M-token GLM-5.3 contexts on 16 H200s with PCP, DCP and Hybrid HiSparse vLLM tools