Search results for "gpu optimization"

FlashAttention-4 Hits 71% GPU Utilization on NVIDIA Blackwell B200

FlashAttention-4 Hits 71% GPU Utilization on NVIDIA Blackwell B200

Together AI's FlashAttention-4 achieves 1,605 TFLOPs/s on B200 GPUs, up to 2.7x faster than Triton. New pipelining overcomes asymmetric hardware scaling bottlenecks.

NVIDIA Advances AI Infrastructure With Disaggregated LLM Inference on Kubernetes

NVIDIA Advances AI Infrastructure With Disaggregated LLM Inference on Kubernetes

NVIDIA details new Kubernetes deployment patterns for disaggregated LLM inference using Dynamo and Grove, promising better GPU utilization for AI workloads.

NVIDIA MIG Boosts AI Infrastructure ROI by 33% Over Time-Slicing

NVIDIA MIG Boosts AI Infrastructure ROI by 33% Over Time-Slicing

New NVIDIA benchmarks show Multi-Instance GPU partitioning achieves 1.00 req/s per GPU versus 0.76 for time-slicing in production AI workloads.

Together AI Kernels Team Achieves 3.6x Performance Gains on NVIDIA Hardware

Together AI Kernels Team Achieves 3.6x Performance Gains on NVIDIA Hardware

Together AI's kernel research team delivers major GPU optimization breakthroughs, cutting inference latency from 281ms to 77ms for enterprise AI deployments.

AMD’s TLX Kernel Boosts GEMM Performance for LLM Training

AMD’s TLX Kernel Boosts GEMM Performance for LLM Training

AMD's TLX optimizations accelerate GEMM, cutting memory bottlenecks and enhancing large language model training on GPUs.

AMD ROCm 10 Debuts with ROCm.AI, Promises 3.3x AI Gains

AMD ROCm 10 Debuts with ROCm.AI, Promises 3.3x AI Gains

AMD launches ROCm 10, marking a 10-year milestone with new AI-native tools like ROCm.AI, delivering 3.3x inference and 2.4x training boosts.

NVIDIA Introduces Reshard Collective to Optimize RL Weight Refit

NVIDIA Introduces Reshard Collective to Optimize RL Weight Refit

NVIDIA's Reshard collective reduces reinforcement learning weight refit latency, addressing inefficiencies in GPU training-inference synchronization.

WordPress Plugin to Boost Copyright Protection by Timestamping Content on the EOSIO Blockchain

WordPress Plugin to Boost Copyright Protection by Timestamping Content on the EOSIO Blockchain

Upon setting up a plugin dubbed WordProof Timestamp, WordPress users will be able to encode their content on EOSIO blockchain, securing copyright protection.

Yearn.finance Takes Pickle Finance Protocol Onboard to Enhance DeFi Rewards

Yearn.finance Takes Pickle Finance Protocol Onboard to Enhance DeFi Rewards

Leading DeFi protocol Yearn.finance has absorbed Pickle Finance to create a structure that will leverage shared expertise, boost specialization, and minimize work duplication.

Denmark Backs Blockchain Technology As Instrument to Fighting Runaway Political Corruption

Denmark Backs Blockchain Technology As Instrument to Fighting Runaway Political Corruption

The Danish government has thrown its support behind blockchain in the fight against political corruption.

How to Build a Decentralized Exchange (DEX) Like Uniswap in Less than One Hour

How to Build a Decentralized Exchange (DEX) Like Uniswap in Less than One Hour

Uniswap is an open-source exchange, it could be easily built by the existing codes that exists on GitHub and etherscan.

ADA Price Surges Over 100% As Cardano Goguen Update Set For March

ADA Price Surges Over 100% As Cardano Goguen Update Set For March

The ADA price surged by over 100% this week and with Cardano’s Goguen update and functional smart contract integration set for March, ADA’s price rally looks far from over