Skip to content

[https://nvbugs/5666804][test] only adding sampler config for limited models #9512

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
LarryXFly merged 7 commits into from
Dec 8, 2025
Merged

[https://nvbugs/5666804][test] only adding sampler config for limited models #9512

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
cb8c823
only adding sampler config for limited models
ruodil Nov 27, 2025
2aa5706
Merge branch 'main' into ruodil/perf
yufeiwu-nv Nov 28, 2025
2f2840e
Merge branch 'main' into ruodil/perf
LarryXFly Dec 3, 2025
7815f33
Merge branch 'main' into ruodil/perf
LarryXFly Dec 3, 2025
49009aa
Merge branch 'main' into ruodil/perf
ruodil Dec 4, 2025
ee4a7a5
Merge branch 'main' into ruodil/perf
ruodil Dec 5, 2025
f16d6f9
Merge branch 'main' into ruodil/perf
ruodil Dec 8, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
20 changes: 15 additions & 5 deletions tests/integration/defs/perf/sampler_options_config.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -26,9 +26,19 @@ def get_sampler_options_config(model_label: str) -> dict:
Returns:
dict: sampler options config
"""
base_config = {
'top_k': 4,
'top_p': 0.5,
'temperature': 0.5,
}
base_config = {}
if model_label in [
'llama_v3.1_70b_instruct-bench-pytorch-bfloat16-maxbs:512-maxnt:2048-input_output_len:200,2000-reqs:64-con:200-gpus:8',
'llama_v3.1_70b_instruct_fp8-bench-pytorch-float8-maxbs:512-maxnt:2048-input_output_len:128,128-gpus:8',
'llama_v3.2_1b-bench-pytorch-bfloat16-maxbs:512-maxnt:2048-input_output_len:500,2000-gpus:2',
'llama_v3.3_70b_instruct_fp8-bench-pytorch-float8-maxbs:512-maxnt:2048-input_output_len:128,128-gpus:4',
'llama_v4_maverick_17b_128e_instruct_fp8-bench-pytorch-float8-maxbs:1024-maxnt:20000-kv_frac:0.6-input_output_len:20000,2000-reqs:1000-ep:8-gpus:8',
'llama_v4_maverick_17b_128e_instruct_fp8-bench-pytorch-float8-maxbs:1024-maxnt:4096-kv_frac:0.85-input_output_len:1000,1000-reqs:3000-ep:8-gpus:8',
'llama_v4_scout_17b_16e_instruct_fp8-bench-pytorch-float8-maxbs:1024-maxnt:4096-kv_frac:0.85-input_output_len:500,2000-reqs:3000-ep:8-gpus:8',
'mistral_small_v3.1_24b-bench-pytorch-bfloat16-maxbs:512-maxnt:2048-input_output_len:1000,2000-reqs:500-con:200-gpus:2',
'phi_4_mini_instruct-bench-pytorch-bfloat16-maxbs:512-maxnt:2048-input_output_len:128,128'
]:
base_config['top_k'] = 4
base_config['top_p'] = 0.5
base_config['temperature'] = 0.5
return base_config
9 changes: 5 additions & 4 deletions tests/integration/defs/perf/test_perf.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1784,10 +1784,11 @@ def get_trtllm_bench_command(self, engine_dir):
if not os.path.exists(sampler_options_path):
os.makedirs(os.path.dirname(sampler_options_path), exist_ok=True)
sampler_config = get_sampler_options_config(self._config.to_string())
print_info(f"sampler options config: {sampler_config}")
with open(sampler_options_path, 'w') as f:
yaml.dump(sampler_config, f, default_flow_style=False)
benchmark_cmd += [f"--sampler_options={sampler_options_path}"]
if sampler_config:
print_info(f"sampler options config: {sampler_config}")
with open(sampler_options_path, 'w') as f:
yaml.dump(sampler_config, f, default_flow_style=False)
benchmark_cmd += [f"--sampler_options={sampler_options_path}"]
return benchmark_cmd

def get_commands(self):
Expand Down
Loading