| .. |
|
all_reduce
|
156f577f79
feat: switch from `PYBIND11_MODULE` to `TORCH_LIBRARY` (#569)
|
vor 7 Monaten |
|
attention
|
156f577f79
feat: switch from `PYBIND11_MODULE` to `TORCH_LIBRARY` (#569)
|
vor 7 Monaten |
|
backup
|
f8dfac6372
chore: attention refactor and upstream sync apr01 (#365)
|
vor 11 Monaten |
|
cpu
|
271a680026
feat: inference support for PowerPC ISA
|
vor 7 Monaten |
|
hadamard
|
5d288aa76c
feat: add fast hadamard transformation kernels (#232)
|
vor 1 Jahr |
|
mamba
|
5be90c3859
Mamba infrastrucuture support (#586)
|
vor 6 Monaten |
|
moe
|
156f577f79
feat: switch from `PYBIND11_MODULE` to `TORCH_LIBRARY` (#569)
|
vor 7 Monaten |
|
punica
|
4f42985b5c
feat: qwen2 lora shapes
|
vor 7 Monaten |
|
quantization
|
ad24e74a99
feat: FP8 weight-only quantization support for Ampere GPUs
|
vor 6 Monaten |
|
activation_kernels.cu
|
c0c336aaa3
refactor: registry for processing model inputs; quick_gelu; clip model support
|
vor 7 Monaten |
|
cache.h
|
156f577f79
feat: switch from `PYBIND11_MODULE` to `TORCH_LIBRARY` (#569)
|
vor 7 Monaten |
|
cache_kernels.cu
|
156f577f79
feat: switch from `PYBIND11_MODULE` to `TORCH_LIBRARY` (#569)
|
vor 7 Monaten |
|
cuda_compat.h
|
00acf371f9
rocm: fused topk softmax
|
vor 7 Monaten |
|
cuda_utils.h
|
156f577f79
feat: switch from `PYBIND11_MODULE` to `TORCH_LIBRARY` (#569)
|
vor 7 Monaten |
|
cuda_utils_kernels.cu
|
156f577f79
feat: switch from `PYBIND11_MODULE` to `TORCH_LIBRARY` (#569)
|
vor 7 Monaten |
|
dispatch_utils.h
|
156f577f79
feat: switch from `PYBIND11_MODULE` to `TORCH_LIBRARY` (#569)
|
vor 7 Monaten |
|
layernorm_kernels.cu
|
156f577f79
feat: switch from `PYBIND11_MODULE` to `TORCH_LIBRARY` (#569)
|
vor 7 Monaten |
|
ops.h
|
5be90c3859
Mamba infrastrucuture support (#586)
|
vor 6 Monaten |
|
pos_encoding_kernels.cu
|
156f577f79
feat: switch from `PYBIND11_MODULE` to `TORCH_LIBRARY` (#569)
|
vor 7 Monaten |
|
reduction.cuh
|
aba03b4756
feat: dynamic per-token activation quantization
|
vor 7 Monaten |
|
registration.h
|
156f577f79
feat: switch from `PYBIND11_MODULE` to `TORCH_LIBRARY` (#569)
|
vor 7 Monaten |
|
torch_bindings.cpp
|
ad24e74a99
feat: FP8 weight-only quantization support for Ampere GPUs
|
vor 6 Monaten |
AlpinDale
ad24e74a99
feat: FP8 weight-only quantization support for Ampere GPUs