Tencent’s tech team has optimized DeepSeek’s open-source DeepEP communication framework,hottest video game sex scenes boosting its performance across different network environments, according to the Chinese AI startup. Testing showed a 100% improvement on RoCE networks and a 30% gain on InfiniBand (IB), offering more efficient solutions for AI model training. On GitHub, DeepSeek acknowledged the Chinese tech giant’s contribution had led to a “huge speedup.” DeepEP is a communication library tailored for a mixture of experts (MoE) and expert parallelism (EP), supporting high-throughput, low-latency GPU kernels and low-precision computing, including FP8. Tencent’s Starlink Networking team identified two main bottlenecks: underutilized dual-port NIC bandwidth and CPU control latency. After targeted optimizations, performance doubled on RoCE and improved by 30% on IB. The enhanced framework is now fully open-source and has been successfully deployed in training Tencent’s Hunyuan large model, demonstrating strong versatility within environments built on Tencent’s Starlink and H20 servers, Chinese tech media outlet iThome reported. [iThome, in Chinese]
(Editor: {typename type="name"/})
Best Apple Pencil Pro deal: Save $30 at Best Buy
5 crucial ways men can help end sexual assault
Woody Allen's response to Harvey Weinstein's behavior is as gross as you'd expect
This fire safety video about smoke alarms is so gloriously odd
Dyson Supersonic deal: Save $100 on the blow dryer
There's a red sun over Britain and Londoners are loving it
You know you want to watch a 1,300
Prisoners in Texas donate more than $50,000 for Harvey relief
Google Pixel brings back popular camera features in new update
Rihanna will soon have a street named after her
接受PR>=1、BR>=1,流量相当,内容相关类链接。