mirror of
https://github.com/NVIDIA/TensorRT-LLM.git
synced 2026-01-13 22:18:36 +08:00
Merge 30a6a40d2d into 6df2c8a074
This commit is contained in:
commit
efe9e376a8
@ -68,8 +68,8 @@ This directory contains various test configuration files:
|
||||
- `llm_triton_integration.txt` - Triton backend integration tests
|
||||
|
||||
### Release-Specific Tests
|
||||
- `llm_digits_func.txt` - Functional tests for DIGITS release
|
||||
- `llm_digits_perf.txt` - Performance tests for DIGITS release
|
||||
- `llm_spark_func.txt` - Functional tests for DGX-Spark release
|
||||
- `llm_spark_perf.yml` - Performance tests for DGX-Spark release
|
||||
|
||||
## Test Execution Schedule
|
||||
|
||||
|
||||
@ -1,16 +1,10 @@
|
||||
llm_digits_perf:
|
||||
version: 0.0.1
|
||||
llm_spark_perf:
|
||||
- condition:
|
||||
ranges:
|
||||
system_gpu_count:
|
||||
gte: 1
|
||||
lte: 1
|
||||
wildcards:
|
||||
gpu:
|
||||
- '*gb10*'
|
||||
linux_distribution_name: ubuntu*
|
||||
cpu: aarch64
|
||||
terms:
|
||||
backend: pytorch
|
||||
tests:
|
||||
- perf/test_perf.py::test_perf[gpt_oss_20b_fp4-bench-pytorch-streaming-float4-maxbs:1-input_output_len:2048,128-reqs:1-con:1]
|
||||
- perf/test_perf.py::test_perf[gpt_oss_120b_fp4-bench-pytorch-streaming-float4-maxbs:1-input_output_len:2048,128-reqs:1-con:1]
|
||||
Loading…
Reference in New Issue
Block a user