Code
Hub
Workspaces
Connect
Indexed graphs
Engine
MCP
copy
hub
/
github.com/NVIDIA/TensorRT-LLM
/ generate_kernels.py
File
generate_kernels.py
cpp/tensorrt_llm/kernels/cutlass_kernels/python/generate_kernels.py:None–None ·
view source on GitHub ↗
Source
from the content-addressed store, hash-verified
1
import
argparse
2
import
enum
3
import
os
4
from
itertools
import
chain, product
Callers
nothing calls this directly
Calls
14
e2m1_type
Class · 0.85
generate_sm120_operations
Function · 0.85
has_arch
Function · 0.85
generate_sm103_operations
Function · 0.85
generate_sm100_operations
Function · 0.85
generate_sm90_operations
Function · 0.85
generate_sm80_operations
Function · 0.85
should_skip
Function · 0.85
is_mixed_dtype_grouped
Function · 0.85
write_file
Function · 0.85
clean_leftover_files
Function · 0.85
split
Method · 0.45
Tested by
no test coverage detected