hgpu.org » nVidia GeForce RTX 3090
Fabian Knorr, Peter Thoman, Thomas Fahringer
Tags: Compression, Computer science, CUDA, nVidia, nVidia GeForce RTX 2070, nVidia GeForce RTX 3090, Package, SYCL, Tesla V100
August 8, 2021 by hgpu
Boyuan Feng, Yuke Wang, Tong Geng, Ang Li, Yufei Ding
Tags: Algorithms, Computer science, CUDA, Deep learning, Neural networks, nVidia, nVidia GeForce RTX 3090, Package, Precision, Python, Tesla A100
June 27, 2021 by hgpu
Recent source codes
* * *
Most viewed papers (last 30 days)
- DeepSeek-V4-Flash on AMD gfx90a: Correctness Recovery and Inference Performance Engineering
- AutoTuneBench: Trustworthy Measurement for Agent Auto-Tuning of LLM Serving Engines
- Accelerating the Solving of Many Tiny General Linear Systems on GPUs: Application to Constitutive Laws
- PrefixBench-H100: Characterizing Prefix Reuse and Time-to-First-Token in H100 LLM Serving
- Hardware-Aware FP4 FlashAttention-4
- Every Kernel Is a Join: Automatic Multi-GPU Parallelism for AI Computations in Einsummable
- Accelerating the Mitigation of LLM Inference Nondeterminism Across GPU Architectures
- pytest-gpu-proof: Enabling Cloud-CPU Continuous Integration for GPU Code with Local GPU Attestation
- Stencil Computation at the Intersection of AI and HPC
- Microarchitectural Memory Bandwidth Saturation, KV-Cache Paging Dynamics, and Time-to-First-Token Latency: A Comparative Benchmark of vLLM, TensorRT-LLM, and FlashAttention-3 on NVIDIA Hopper H100 versus AMD Instinct MI300X
* * *



