34 skills found · Page 1 of 2
666DZY666 / Micronetmicronet, a model compression and deploy lib. compression: 1、quantization: quantization-aware-training(QAT), High-Bit(>2b)(DoReFa/Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inference)、Low-Bit(≤2b)/Ternary and Binary(TWN/BNN/XNOR-Net); post-training-quantization(PTQ), 8-bit(tensorrt); 2、 pruning: normal、regular and group convolutional channel pruning; 3、 group convolution structure; 4、batch-normalization fuse for quantization. deploy: tensorrt, fp32/fp16/int8(ptq-calibration)、op-adapt(upsample)、dynamic_shape
salykova / Sgemm.cMulti-Threaded FP32 Matrix Multiplication on x86 CPUs
tgautam03 / XGeMMAccelerated General (FP32) Matrix Multiplication from scratch in CUDA
salykova / Sgemm.cuHigh-Performance FP32 GEMM on CUDA devices
seb-v / Fp32 Sgemm AmdSuper fast FP32 matrix multiplication on RDNA3
ratszhu / Z Image Turbo Carto是一个专为高性能本地部署设计的高级 WebUI 工程,支持Mac/Windows本地部署，本项目基于阿里通义 Tongyi-MAI/Z-Image-Turbo 模型，针对 Apple Silicon (M1/M2/M3/M4) 和 NVIDIA RTX (Windows) 进行了深度底层优化。通过 Bfloat16 精度、VAE FP32 混合精度推理以及手动 LoRA 注入技术，实现了极速、高清、低显存占用的完美平衡。
kentaroy47 / Benchmark FP32 FP16 INT8 With TensorRTBenchmark inference speed of CNNs with various quantization methods in Pytorch+TensorRT with Jetson Nano/Xavier
JohndeVostok / APEA GPU FP32 computation method with Tensor Cores.
gigit0000 / Qwen3.cLightweight C inference for Qwen3 GGUF. Multiturn prefix caching & batch processing.
aditya4d / Gemm Vega64Implement asm gemm on vega64 for 4096x4096 fp32 matrix
yhinai / TensorGPGPURISC-V vector and tensor compute extensions for Vortex GPGPU acceleration for ML workloads. Optimized for transformer models, CNNs, and generative AI with configurable precision (FP32/16/BF16/INT8).
enp1s0 / CuMpSGEMMFast SGEMM emulation on Tensor Cores
hukenovs / Fp32 LogicFloating point FP32 core HDL. For Xilinx FPGAs. Include base converters and some math functions.
minhsun-c / FP32 Matrix MultiplierHigh-performance systolic-array accelerator for FP32 matrix multiplication in deep learning.
gongouveia / Resnet Quantization ExperimentsTools for per layer quantization, fp32, fp16 , PTQ and QAT int8 (int4 not yet implemented)
zeng-zuoqi / Cublas Gemm Benchmarkcublas gemm benchmark (fp32 fp16 int8 fp16(tensor core) int8(tensor core))
PixWizardry / ComfyUI Z Image FP32No description available
IFeelBloated / Warpsharpfp32 warpsharp for vaporsynth
Puppetmaster134 / Fp ConverterConvert Pytorch FP32, FP16, and BFloat16 to FP8 and back again
Const-me / CbrtPsA function to compute FP32 cubic root with SIMD on PCs
