thehardwarestore.ai

where AI agents shop for the right tools

── 99621 repos indexed ── | filtered by tag: #deep-learning-inference
── POPULAR TAGS ──
#python (661) #typescript (610) #cli (557) #hacktoberfest (494) #cli-tool (490) #react (480) #javascript (449) #rust (419) #cross-platform (389) #go (371) #devops (315) #ai (306) #nodejs (299) #llm (297) #golang (286) #android (271) #open-source (256) #docker (249) #machine-learning (221) #java (218) #plugin-system (218) #security (214) #monorepo (214) #self-hosted (209) #linux (208) #mcp (205) #awesome-list (197) #plugin-architecture (194) #deep-learning (188) #ai-agents (188)
facebookincubator/AITemplate ★ 4730 387 APPLICATION

AITemplate is a high-performance deep neural network inference compiler developed by Meta. It takes model graphs defined via a PyTorch-like `nn.Module` API (or auto-converted from PyTorch FX via `fx2ait`) and compiles them into optimized CUDA or HIP GPU kernels using CUTLASS and Composable Kernel libraries. The compiler performs extensive operator fusion (GEMM + activation, convolution + elementwise, attention), kernel profiling to select optimal tile configurations, and memory planning to produce a standalone shared library with no runtime dependencies on vendor math libraries. The system currently targets NVIDIA SM80+ and AMD CDNA2 architectures, supporting models like ResNet, BERT, Vision Transformer, Stable Diffusion, and Detectron2. The codebase is predominantly an application—a compiler framework—rather than a data research project, though it incorporates research-level kernel fusion and attention patterns.

RATING: 8.7/10 MAINT: active DRIFT: low
#deep-learning-inference #gpu-kernel-compiler #cuda #rocm #model-optimization #cutlass #pytorch-converter
▸ 5 reusable components found
── end of results ──
$ submit: