Skip to content

Update benchmark run configuration and lite profile #533

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
DhavalRepo18 merged 1 commit into from
Sep 7, 2026
Merged

Update benchmark run configuration and lite profile #533

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
63 changes: 50 additions & 13 deletions benchmarks/run.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,27 +2,64 @@

set -euo pipefail

scenario_dir="${1:-${SCENARIO_DIR:-}}"
leaderboard_dir="${2:-${LEADERBOARD_DIR:-}}"
usage() {
printf 'Usage: %s -s SCENARIO_DIR -l LEADERBOARD_DIR -t TRAJECTORY_DIR\n' "$0" >&2
printf ' %s SCENARIO_DIR LEADERBOARD_DIR TRAJECTORY_DIR\n' "$0" >&2
printf ' or set SCENARIO_DIR, LEADERBOARD_DIR, and TRAJECTORY_DIR\n' >&2
}

if [[ -z "$scenario_dir" || -z "$leaderboard_dir" ]]; then
printf 'Usage: %s SCENARIO_DIR LEADERBOARD_DIR\n' "$0" >&2
printf ' or set SCENARIO_DIR and LEADERBOARD_DIR\n' >&2
scenario_dir="${SCENARIO_DIR:-}"
leaderboard_dir="${LEADERBOARD_DIR:-}"
trajectory_dir="${TRAJECTORY_DIR:-}"

while getopts ':s:l:t:' option; do
case "$option" in
s) scenario_dir="$OPTARG" ;;
l) leaderboard_dir="$OPTARG" ;;
t) trajectory_dir="$OPTARG" ;;
:) printf 'Option -%s requires an argument.\n' "$OPTARG" >&2; usage; exit 2 ;;
\?) printf 'Unknown option: -%s\n' "$OPTARG" >&2; usage; exit 2 ;;
esac
done

option_arg_count=$((OPTIND - 1))
shift "$option_arg_count"

if (( option_arg_count > 0 && $# > 0 )); then
printf 'Options and positional arguments cannot be combined.\n' >&2
usage
exit 2
fi

if (( option_arg_count == 0 )); then
if (( $# > 3 )); then
usage
exit 2
fi
scenario_dir="${1:-$scenario_dir}"
leaderboard_dir="${2:-$leaderboard_dir}"
trajectory_dir="${3:-$trajectory_dir}"
fi

if [[ -z "$scenario_dir" || -z "$leaderboard_dir" || -z "$trajectory_dir" ]]; then
usage
exit 2
fi

agent_name=stirrup_agent
scenario_ids=lite
scenario_ids=tsfm_lite

model_configs=(
"litellm_proxy/gcp/gemini-3.6-flash high"
"litellm_proxy/azure/gpt-5.6-sol max"
"tokenrouter/z-ai/glm-5.2 max"
"tokenrouter/MiniMax-M3 high"
"litellm_proxy/aws/claude-opus-5 high"
"litellm_proxy/azure/DeepSeek-V4-Flash max"
"litellm_proxy/aws/gpt-oss-120b high"
"litellm_proxy/aws/claude-sonnet-5 max"
"tokenrouter/MiniMax-M3 high"
"tokenrouter/moonshotai/kimi-k3 max"
"tokenrouter/z-ai/glm-5.3 max"
"tokenrouter/deepseek/deepseek-v4-flash max"
""

)

for model_config in "${model_configs[@]}"; do
Expand All @@ -36,11 +73,11 @@ for model_config in "${model_configs[@]}"; do
--agent_name "$agent_name" \
--model-id "$model_id" \
--reasoning-effort "$reasoning_effort" \
--trajectory-root "$leaderboard_dir/assetopsbench-trajectories" \
--trajectory-root "$trajectory_dir" \
--reports-root "$leaderboard_dir/assetopsbench-reports" \
--stirrup-workspace-root "$leaderboard_dir/assetopsbench-stirrup-workspaces" \
--skip-existing \
--continue-on-error \
--preserve-workspaces
# --continue-on-error
# --skip-existing \

done
48 changes: 37 additions & 11 deletions benchmarks/scenario_suite/lite.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,35 +2,36 @@ car:
- 151
- 152
- 153
- 156
- 167
- 157
- 159
- 160
- 178
- 180
- 182
- 183
- 189
- 193
fcc:
- 301
- 303
- 304
- 305
- 308
- 314
- 316
- 312
- 320
- 323
- 322
- 325
- 327
fmsr:
- 902
- 904
- 905
- 906
- 908
- 915
- 916
- 917
- 920
- 923
- 928
- 932
health:
- 401
- 402
Expand All @@ -42,11 +43,36 @@ health:
- 408
- 409
- 410
tsfm: []
tsfm:
- 1001
- 1002
- 1003
- 1004
- 1005
- 1006
- 1007
- 1008
- 1009
- 1010
- 1011
- 1012
- 1013
- 1014
- 1015
- 1016
- 1017
- 1018
- 1019
- 1020
- 1021
- 1022
- 1023
- 1024
- 1025
wosr:
- 5
- 9
- 11
- 13
- 16
- 20
- 24
- 31
Expand Down