.gitignore
.gitlab-ci.yml
.pre-commit-config.yaml
Dockerfile.test
LICENSE
Makefile
README.md
dockerfile
dockerfile.helper
pyproject.toml
tox.ini
versions.json
dai_bench/__init__.py
dai_bench/__version__.py
dai_bench/logging.py
dai_bench.egg-info/PKG-INFO
dai_bench.egg-info/SOURCES.txt
dai_bench.egg-info/dependency_links.txt
dai_bench.egg-info/entry_points.txt
dai_bench.egg-info/requires.txt
dai_bench.egg-info/scm_file_list.json
dai_bench.egg-info/scm_version.json
dai_bench.egg-info/top_level.txt
dai_bench/_vendor/config_explorer/__init__.py
dai_bench/_vendor/config_explorer/capacity_planner.py
dai_bench/_vendor/config_explorer/constants.py
dai_bench/_vendor/config_explorer/convert.py
dai_bench/_vendor/config_explorer/explorer.py
dai_bench/_vendor/config_explorer/plotting.py
dai_bench/_vendor/config_explorer/schema.py
dai_bench/_vendor/config_explorer/vendor.lock
dai_bench/_vendor_shims/config_explorer.py
dai_bench/cli/__init__.py
dai_bench/cli/main.py
dai_bench/cli/run.py
dai_bench/cli/setup.py
dai_bench/core/__init__.py
dai_bench/core/config.py
dai_bench/core/launch.py
dai_bench/core/prism_report.py
dai_bench/core/runner.py
dai_bench/runners/aiperf_bench.py
dai_bench/runners/base.py
dai_bench/runners/vllm_bench.py
docs/api.md
docs/index.md
docs/tutorial/baize-analysis.png
docs/tutorial/baize-dataset.jpg
docs/tutorial/baize-notebook-ssh.png
docs/tutorial/baize-notebook.jpg
docs/tutorial/config-explorer.md
docs/tutorial/config-explorer.png
docs/tutorial/config-performance.png
docs/tutorial/jupyter.md
docs/tutorial/k8s-server.md
docs/tutorial/k8s.md
docs/tutorial/llm-d-prism.md
docs/tutorial/local.md
docs/tutorial/performance-tradeoff.png
docs/tutorial/quickstart.md
examples/README.md
examples/data/deepseek-v3.2-pd-disagg/results/01/_gpfs_rd_models_DeepSeek-V3.2/data/stage1-c16_32_64_128-i100-o100/benchmark_report,_infqps-concurrency128-20260112-030906.json.yaml
examples/data/deepseek-v3.2-pd-disagg/results/01/_gpfs_rd_models_DeepSeek-V3.2/data/stage1-c16_32_64_128-i100-o100/benchmark_report,_infqps-concurrency16-20260112-030744.json.yaml
examples/data/deepseek-v3.2-pd-disagg/results/01/_gpfs_rd_models_DeepSeek-V3.2/data/stage1-c16_32_64_128-i100-o100/benchmark_report,_infqps-concurrency32-20260112-030832.json.yaml
examples/data/deepseek-v3.2-pd-disagg/results/01/_gpfs_rd_models_DeepSeek-V3.2/data/stage1-c16_32_64_128-i100-o100/benchmark_report,_infqps-concurrency64-20260112-030852.json.yaml
examples/data/deepseek-v3.2-pd-disagg/results/01/_gpfs_rd_models_DeepSeek-V3.2/data/stage1-c16_32_64_128-i100-o100/infqps-concurrency128-20260112-030906.json
examples/data/deepseek-v3.2-pd-disagg/results/01/_gpfs_rd_models_DeepSeek-V3.2/data/stage1-c16_32_64_128-i100-o100/infqps-concurrency16-20260112-030744.json
examples/data/deepseek-v3.2-pd-disagg/results/01/_gpfs_rd_models_DeepSeek-V3.2/data/stage1-c16_32_64_128-i100-o100/infqps-concurrency32-20260112-030832.json
examples/data/deepseek-v3.2-pd-disagg/results/01/_gpfs_rd_models_DeepSeek-V3.2/data/stage1-c16_32_64_128-i100-o100/infqps-concurrency64-20260112-030852.json
examples/data/deepseek-v3.2-pd-disagg/results/01/_gpfs_rd_models_DeepSeek-V3.2/data/stage1-c16_32_64_128-i4000-o100/benchmark_report,_infqps-concurrency128-20260112-032637.json.yaml
examples/data/deepseek-v3.2-pd-disagg/results/01/_gpfs_rd_models_DeepSeek-V3.2/data/stage1-c16_32_64_128-i4000-o100/benchmark_report,_infqps-concurrency16-20260112-032327.json.yaml
examples/data/deepseek-v3.2-pd-disagg/results/01/_gpfs_rd_models_DeepSeek-V3.2/data/stage1-c16_32_64_128-i4000-o100/benchmark_report,_infqps-concurrency32-20260112-032445.json.yaml
examples/data/deepseek-v3.2-pd-disagg/results/01/_gpfs_rd_models_DeepSeek-V3.2/data/stage1-c16_32_64_128-i4000-o100/benchmark_report,_infqps-concurrency64-20260112-032541.json.yaml
examples/data/deepseek-v3.2-pd-disagg/results/01/_gpfs_rd_models_DeepSeek-V3.2/data/stage1-c16_32_64_128-i4000-o100/infqps-concurrency128-20260112-032637.json
examples/data/deepseek-v3.2-pd-disagg/results/01/_gpfs_rd_models_DeepSeek-V3.2/data/stage1-c16_32_64_128-i4000-o100/infqps-concurrency16-20260112-032327.json
examples/data/deepseek-v3.2-pd-disagg/results/01/_gpfs_rd_models_DeepSeek-V3.2/data/stage1-c16_32_64_128-i4000-o100/infqps-concurrency32-20260112-032445.json
examples/data/deepseek-v3.2-pd-disagg/results/01/_gpfs_rd_models_DeepSeek-V3.2/data/stage1-c16_32_64_128-i4000-o100/infqps-concurrency64-20260112-032541.json
examples/data/deepseek-v3.2-pd-disagg/results/01/_gpfs_rd_models_DeepSeek-V3.2/data/stage2-c16_32_64_128-i10000-o100/benchmark_report,_infqps-concurrency128-20260112-031616.json.yaml
examples/data/deepseek-v3.2-pd-disagg/results/01/_gpfs_rd_models_DeepSeek-V3.2/data/stage2-c16_32_64_128-i10000-o100/benchmark_report,_infqps-concurrency16-20260112-030917.json.yaml
examples/data/deepseek-v3.2-pd-disagg/results/01/_gpfs_rd_models_DeepSeek-V3.2/data/stage2-c16_32_64_128-i10000-o100/benchmark_report,_infqps-concurrency32-20260112-031137.json.yaml
examples/data/deepseek-v3.2-pd-disagg/results/01/_gpfs_rd_models_DeepSeek-V3.2/data/stage2-c16_32_64_128-i10000-o100/benchmark_report,_infqps-concurrency64-20260112-031356.json.yaml
examples/data/deepseek-v3.2-pd-disagg/results/01/_gpfs_rd_models_DeepSeek-V3.2/data/stage2-c16_32_64_128-i10000-o100/infqps-concurrency128-20260112-031616.json
examples/data/deepseek-v3.2-pd-disagg/results/01/_gpfs_rd_models_DeepSeek-V3.2/data/stage2-c16_32_64_128-i10000-o100/infqps-concurrency16-20260112-030917.json
examples/data/deepseek-v3.2-pd-disagg/results/01/_gpfs_rd_models_DeepSeek-V3.2/data/stage2-c16_32_64_128-i10000-o100/infqps-concurrency32-20260112-031137.json
examples/data/deepseek-v3.2-pd-disagg/results/01/_gpfs_rd_models_DeepSeek-V3.2/data/stage2-c16_32_64_128-i10000-o100/infqps-concurrency64-20260112-031356.json
examples/data/deepseek-v3.2-pd-disagg/results/02/_gpfs_rd_models_DeepSeek-V3.2/data/stage1-c16_32_64_128-i100-o100/benchmark_report,_infqps-concurrency128-20260112-100231.json.yaml
examples/data/deepseek-v3.2-pd-disagg/results/02/_gpfs_rd_models_DeepSeek-V3.2/data/stage1-c16_32_64_128-i100-o100/benchmark_report,_infqps-concurrency16-20260112-100055.json.yaml
examples/data/deepseek-v3.2-pd-disagg/results/02/_gpfs_rd_models_DeepSeek-V3.2/data/stage1-c16_32_64_128-i100-o100/benchmark_report,_infqps-concurrency32-20260112-100148.json.yaml
examples/data/deepseek-v3.2-pd-disagg/results/02/_gpfs_rd_models_DeepSeek-V3.2/data/stage1-c16_32_64_128-i100-o100/benchmark_report,_infqps-concurrency64-20260112-100210.json.yaml
examples/data/deepseek-v3.2-pd-disagg/results/02/_gpfs_rd_models_DeepSeek-V3.2/data/stage1-c16_32_64_128-i100-o100/infqps-concurrency128-20260112-100231.json
examples/data/deepseek-v3.2-pd-disagg/results/02/_gpfs_rd_models_DeepSeek-V3.2/data/stage1-c16_32_64_128-i100-o100/infqps-concurrency16-20260112-100055.json
examples/data/deepseek-v3.2-pd-disagg/results/02/_gpfs_rd_models_DeepSeek-V3.2/data/stage1-c16_32_64_128-i100-o100/infqps-concurrency32-20260112-100148.json
examples/data/deepseek-v3.2-pd-disagg/results/02/_gpfs_rd_models_DeepSeek-V3.2/data/stage1-c16_32_64_128-i100-o100/infqps-concurrency64-20260112-100210.json
examples/data/deepseek-v3.2-pd-disagg/results/02/_gpfs_rd_models_DeepSeek-V3.2/data/stage2-c16_32_64_128-i4000-o100/benchmark_report,_infqps-concurrency128-20260112-100534.json.yaml
examples/data/deepseek-v3.2-pd-disagg/results/02/_gpfs_rd_models_DeepSeek-V3.2/data/stage2-c16_32_64_128-i4000-o100/benchmark_report,_infqps-concurrency16-20260112-100247.json.yaml
examples/data/deepseek-v3.2-pd-disagg/results/02/_gpfs_rd_models_DeepSeek-V3.2/data/stage2-c16_32_64_128-i4000-o100/benchmark_report,_infqps-concurrency32-20260112-100343.json.yaml
examples/data/deepseek-v3.2-pd-disagg/results/02/_gpfs_rd_models_DeepSeek-V3.2/data/stage2-c16_32_64_128-i4000-o100/benchmark_report,_infqps-concurrency64-20260112-100438.json.yaml
examples/data/deepseek-v3.2-pd-disagg/results/02/_gpfs_rd_models_DeepSeek-V3.2/data/stage2-c16_32_64_128-i4000-o100/infqps-concurrency128-20260112-100534.json
examples/data/deepseek-v3.2-pd-disagg/results/02/_gpfs_rd_models_DeepSeek-V3.2/data/stage2-c16_32_64_128-i4000-o100/infqps-concurrency16-20260112-100247.json
examples/data/deepseek-v3.2-pd-disagg/results/02/_gpfs_rd_models_DeepSeek-V3.2/data/stage2-c16_32_64_128-i4000-o100/infqps-concurrency32-20260112-100343.json
examples/data/deepseek-v3.2-pd-disagg/results/02/_gpfs_rd_models_DeepSeek-V3.2/data/stage2-c16_32_64_128-i4000-o100/infqps-concurrency64-20260112-100438.json
examples/data/deepseek-v3.2-pd-disagg/results/02/_gpfs_rd_models_DeepSeek-V3.2/data/stage3-c16_32_64_128-i10000-o100/benchmark_report,_infqps-concurrency128-20260112-060359.json.yaml
examples/data/deepseek-v3.2-pd-disagg/results/02/_gpfs_rd_models_DeepSeek-V3.2/data/stage3-c16_32_64_128-i10000-o100/benchmark_report,_infqps-concurrency128-20260112-101319.json.yaml
examples/data/deepseek-v3.2-pd-disagg/results/02/_gpfs_rd_models_DeepSeek-V3.2/data/stage3-c16_32_64_128-i10000-o100/benchmark_report,_infqps-concurrency16-20260112-100631.json.yaml
examples/data/deepseek-v3.2-pd-disagg/results/02/_gpfs_rd_models_DeepSeek-V3.2/data/stage3-c16_32_64_128-i10000-o100/benchmark_report,_infqps-concurrency32-20260112-100845.json.yaml
examples/data/deepseek-v3.2-pd-disagg/results/02/_gpfs_rd_models_DeepSeek-V3.2/data/stage3-c16_32_64_128-i10000-o100/benchmark_report,_infqps-concurrency64-20260112-060138.json.yaml
examples/data/deepseek-v3.2-pd-disagg/results/02/_gpfs_rd_models_DeepSeek-V3.2/data/stage3-c16_32_64_128-i10000-o100/benchmark_report,_infqps-concurrency64-20260112-101059.json.yaml
examples/data/deepseek-v3.2-pd-disagg/results/02/_gpfs_rd_models_DeepSeek-V3.2/data/stage3-c16_32_64_128-i10000-o100/infqps-concurrency128-20260112-060359.json
examples/data/deepseek-v3.2-pd-disagg/results/02/_gpfs_rd_models_DeepSeek-V3.2/data/stage3-c16_32_64_128-i10000-o100/infqps-concurrency128-20260112-101319.json
examples/data/deepseek-v3.2-pd-disagg/results/02/_gpfs_rd_models_DeepSeek-V3.2/data/stage3-c16_32_64_128-i10000-o100/infqps-concurrency16-20260112-100631.json
examples/data/deepseek-v3.2-pd-disagg/results/02/_gpfs_rd_models_DeepSeek-V3.2/data/stage3-c16_32_64_128-i10000-o100/infqps-concurrency32-20260112-100845.json
examples/data/deepseek-v3.2-pd-disagg/results/02/_gpfs_rd_models_DeepSeek-V3.2/data/stage3-c16_32_64_128-i10000-o100/infqps-concurrency64-20260112-060138.json
examples/data/deepseek-v3.2-pd-disagg/results/02/_gpfs_rd_models_DeepSeek-V3.2/data/stage3-c16_32_64_128-i10000-o100/infqps-concurrency64-20260112-101059.json
examples/data/deepseek-v32-benchmark/results/03/4_4_EP_EPLB/server.log
examples/data/deepseek-v32-benchmark/results/03/4_4_EP_EPLB/data/stage1-c32_64_128-i100-o1/benchmark_report,_infqps-concurrency128-20260108-122620.json.yaml
examples/data/deepseek-v32-benchmark/results/03/4_4_EP_EPLB/data/stage1-c32_64_128-i100-o1/benchmark_report,_infqps-concurrency32-20260108-122617.json.yaml
examples/data/deepseek-v32-benchmark/results/03/4_4_EP_EPLB/data/stage1-c32_64_128-i100-o1/benchmark_report,_infqps-concurrency64-20260108-122618.json.yaml
examples/data/deepseek-v32-benchmark/results/03/4_4_EP_EPLB/data/stage1-c32_64_128-i100-o1/infqps-concurrency128-20260108-122620.json
examples/data/deepseek-v32-benchmark/results/03/4_4_EP_EPLB/data/stage1-c32_64_128-i100-o1/infqps-concurrency32-20260108-122617.json
examples/data/deepseek-v32-benchmark/results/03/4_4_EP_EPLB/data/stage1-c32_64_128-i100-o1/infqps-concurrency64-20260108-122618.json
examples/data/deepseek-v32-benchmark/results/03/4_4_EP_EPLB/data/stage2-c32_64_128-i10000-o1/benchmark_report,_infqps-concurrency128-20260108-122640.json.yaml
examples/data/deepseek-v32-benchmark/results/03/4_4_EP_EPLB/data/stage2-c32_64_128-i10000-o1/benchmark_report,_infqps-concurrency32-20260108-122621.json.yaml
examples/data/deepseek-v32-benchmark/results/03/4_4_EP_EPLB/data/stage2-c32_64_128-i10000-o1/benchmark_report,_infqps-concurrency64-20260108-122631.json.yaml
examples/data/deepseek-v32-benchmark/results/03/4_4_EP_EPLB/data/stage2-c32_64_128-i10000-o1/infqps-concurrency128-20260108-122640.json
examples/data/deepseek-v32-benchmark/results/03/4_4_EP_EPLB/data/stage2-c32_64_128-i10000-o1/infqps-concurrency32-20260108-122621.json
examples/data/deepseek-v32-benchmark/results/03/4_4_EP_EPLB/data/stage2-c32_64_128-i10000-o1/infqps-concurrency64-20260108-122631.json
examples/data/deepseek-v32-benchmark/results/03/4_4_EP_NA/server.log
examples/data/deepseek-v32-benchmark/results/03/4_4_EP_NA/data/stage1-c32_64_128-i100-o1/benchmark_report,_infqps-concurrency128-20260108-122541.json.yaml
examples/data/deepseek-v32-benchmark/results/03/4_4_EP_NA/data/stage1-c32_64_128-i100-o1/benchmark_report,_infqps-concurrency32-20260108-122535.json.yaml
examples/data/deepseek-v32-benchmark/results/03/4_4_EP_NA/data/stage1-c32_64_128-i100-o1/benchmark_report,_infqps-concurrency64-20260108-122539.json.yaml
examples/data/deepseek-v32-benchmark/results/03/4_4_EP_NA/data/stage1-c32_64_128-i100-o1/infqps-concurrency128-20260108-122541.json
examples/data/deepseek-v32-benchmark/results/03/4_4_EP_NA/data/stage1-c32_64_128-i100-o1/infqps-concurrency32-20260108-122535.json
examples/data/deepseek-v32-benchmark/results/03/4_4_EP_NA/data/stage1-c32_64_128-i100-o1/infqps-concurrency64-20260108-122539.json
examples/data/deepseek-v32-benchmark/results/03/4_4_EP_NA/data/stage2-c32_64_128-i10000-o1/benchmark_report,_infqps-concurrency128-20260108-122602.json.yaml
examples/data/deepseek-v32-benchmark/results/03/4_4_EP_NA/data/stage2-c32_64_128-i10000-o1/benchmark_report,_infqps-concurrency32-20260108-122542.json.yaml
examples/data/deepseek-v32-benchmark/results/03/4_4_EP_NA/data/stage2-c32_64_128-i10000-o1/benchmark_report,_infqps-concurrency64-20260108-122551.json.yaml
examples/data/deepseek-v32-benchmark/results/03/4_4_EP_NA/data/stage2-c32_64_128-i10000-o1/infqps-concurrency128-20260108-122602.json
examples/data/deepseek-v32-benchmark/results/03/4_4_EP_NA/data/stage2-c32_64_128-i10000-o1/infqps-concurrency32-20260108-122542.json
examples/data/deepseek-v32-benchmark/results/03/4_4_EP_NA/data/stage2-c32_64_128-i10000-o1/infqps-concurrency64-20260108-122551.json
examples/data/deepseek-v32-benchmark/results/03/8_2_NA_NA/server.log
examples/data/deepseek-v32-benchmark/results/03/8_2_NA_NA/data/stage1-c32_64_128-i100-o1/benchmark_report,_infqps-concurrency128-20260108-123108.json.yaml
examples/data/deepseek-v32-benchmark/results/03/8_2_NA_NA/data/stage1-c32_64_128-i100-o1/benchmark_report,_infqps-concurrency32-20260108-123049.json.yaml
examples/data/deepseek-v32-benchmark/results/03/8_2_NA_NA/data/stage1-c32_64_128-i100-o1/benchmark_report,_infqps-concurrency64-20260108-123105.json.yaml
examples/data/deepseek-v32-benchmark/results/03/8_2_NA_NA/data/stage1-c32_64_128-i100-o1/infqps-concurrency128-20260108-123108.json
examples/data/deepseek-v32-benchmark/results/03/8_2_NA_NA/data/stage1-c32_64_128-i100-o1/infqps-concurrency32-20260108-123049.json
examples/data/deepseek-v32-benchmark/results/03/8_2_NA_NA/data/stage1-c32_64_128-i100-o1/infqps-concurrency64-20260108-123105.json
examples/k8s/README.md
examples/k8s/config.yaml
examples/k8s/job.yaml
examples/k8s/pvc.yaml
examples/k8s-serve/README.md
examples/k8s-serve/bench.yaml
examples/k8s-serve/run.sh
examples/k8s-serve/serve.yaml
examples/k8s-serve/stop.sh
examples/k8s-serve/lws-deepseek-v3.2/README.md
examples/k8s-serve/lws-deepseek-v3.2/apply.sh
examples/k8s-serve/lws-deepseek-v3.2/config.yaml
examples/k8s-serve/lws-deepseek-v3.2/configmap.yaml
examples/k8s-serve/lws-deepseek-v3.2/delete.sh
examples/k8s-serve/lws-deepseek-v3.2/job.yaml
examples/k8s-serve/lws-deepseek-v3.2/pvc.yaml
examples/k8s-serve/lws-deepseek-v3.2/rbac.yaml
examples/local/README.md
examples/local/bench-multiturn.yaml
examples/local/bench.yaml
examples/local/run.sh
examples/local/serve.yaml
examples/local/stop.sh
hack/cleanup_old_benchmark_files.py
hack/download_tokenizers.py
hack/trivy.sh
hack/util.sh
hack/release/get-version.sh
hack/release/make-branch.sh
hack/release/make-candidate.sh
hack/release/make-release.sh
hack/release/release.sh
release-notes/README.md
release-notes/template.yaml
release-notes/notes/v0.1.0/feature.yaml
requirements/dev.txt
tests/README.md
tests/__init__.py
tests/test_aiperf_bench.py
tests/test_config.py
tests/test_e2e_run_init.py
tests/test_from_treatment.py
tests/test_prism_report.py
tests/test_vllm_bench.py
tests/util.sh
tests/config/host.yaml
tests/config/local-path.yaml
tests/config/run.sh
tests/config/serve.yaml
tests/e2e/k8s_client.sh
tests/e2e/local_client.sh
tests/e2e/local_server.sh
tests/e2e/run_e2e.sh
tests/e2e/shell_checks.sh
tools/gen-release-notes/README.md
tools/gen-release-notes/go.mod
tools/gen-release-notes/go.sum
tools/gen-release-notes/main.go
tools/gen-release-notes/note.go
tools/gen-release-notes/release_notes_schema.json
tools/gen-release-notes/template.go
tools/gen-release-notes/templates/minorReleaseNotes.md
tools/gen-release-notes/templates/releaseNotes.md
tools/gen-release-notes/templates/upgradeNotes.md