gcloud container ai profiles benchmarks list - list benchmarks for a given model and model server

gcloud container ai profiles benchmarks list--model=MODEL--model-server=MODEL_SERVER[--format=FORMAT][--instance-type=INSTANCE_TYPE][--model-server-version=MODEL_SERVER_VERSION][--pricing-model=PRICING_MODEL][--serving-stack=SERVING_STACK][--serving-stack-version=SERVING_STACK_VERSION][--use-case=USE_CASE][--filter=EXPRESSION][--limit=LIMIT][--page-size=PAGE_SIZE][--sort-by=[FIELD,…]][--uri][GCLOUD_WIDE_FLAG …]

For examples of visualizing the benchmarking data, see the accompanying Colabnotebook:https://colab.research.google.com/github/GoogleCloudPlatform/kubernetes-engine-samples/blob/main/ai-ml/notebooks/giq_visualizations.ipynb

--model=MODEL

The model.

--model-server=MODEL_SERVER

The model server.

--format=FORMAT

The format to print the output in. Default is csvprofile, which displays theprofile information in a CSV format, includingcost conversions.

--instance-type=INSTANCE_TYPE

The instance type. If not specified, this defaults to anyinstance type.

--model-server-version=MODEL_SERVER_VERSION

The model server version. Default is latest. Other options include the modelserver version of a profile, all which returns all versions.

--pricing-model=PRICING_MODEL

The pricing model to use to calculate token cost. Currently, this supportson-demand, spot, 3-years-cud, 1-year-cud

--serving-stack=SERVING_STACK

The serving stack to filter benchmarking data by. If not provided, benchmarkingdata for all serving stacks that support the given model and model server willbe returned.

--serving-stack-version=SERVING_STACK_VERSION

The serving stack version to filter benchmarking data by. If not provided,benchmarking data for all versions that support the given model and model serverwill be returned.

--use-case=USE_CASE

If specified, results will only show profiles that match the provided use case.Options are: Advanced Customer Support, Code Completion, Text Summarization,Chatbot (ShareGPT), Code Generation, Deep Research.

--filter=EXPRESSION

Apply a Boolean filterEXPRESSION to each resource itemto be listed. If the expression evaluatesTrue, then that item islisted. For more details and examples of filter expressions, run $gcloud topic filters. This flaginteracts with other flags that are applied in this order:--flatten,--sort-by,--filter,--limit.

--limit=LIMIT

Maximum number of resources to list. The default isunlimited. Thisflag interacts with other flags that are applied in this order:--flatten,--sort-by,--filter,--limit.

--page-size=PAGE_SIZE

Some services group resource list output into pages. This flag specifies themaximum number of resources per page. The default is determined by the serviceif it supports paging, otherwise it isunlimited (no paging).Paging may be applied before or after--filter and--limit depending on the service.

--sort-by=[FIELD,…]

Comma-separated list of resource field key names to sort by. The default orderis ascending. Prefix a field with ``~´´ for descending order on thatfield. This flag interacts with other flags that are applied in this order:--flatten,--sort-by,--filter,--limit.

--uri

Print a list of resource URIs instead of the default output, and change thecommand output to a list of URIs. If this flag is used with--format, the formatting is applied on this URI list. To displayURIs alongside other keys instead, use theuri() transform.

Run$gcloud help for details.