# Admin ## co admin Admin operations **Options** ``` -h, --help help for admin ``` ## co admin baselines Baseline management **Options** ``` -h, --help help for baselines ``` ## co admin baselines get get baseline state from ID ``` co admin baselines get [flags] ``` **Options** ``` -h, --help help for get --id string id of finetune --models get information about models deployed in this state only --output string output path, default stdout (default "-") ``` ## co admin baselines list list baseline states ``` co admin baselines list [flags] ``` **Options** ``` --after string retrieve states that were set after this date in dd/mm/yyyy format --before string retrieve states that were set before this date in dd/mm/yyyy format --email string email of deployer -h, --help help for list --id string id of finetune --models get information about models deployed in this state only --num-states int number of states to return (default 10) --order string ordering by deployed time (default "desc") --output string output path, default stdout (default "-") --show-id get id of baseline states ``` ## co admin baselines update update baseline state ``` co admin baselines update [flags] ``` **Options** ``` --force force deployment -h, --help help for update --state string path to state file, or read from stdin by specifying "-" --vx-yaml string path to vx yaml, or read from stdin by specifying "-" ``` ## co admin limits **Options** ``` -h, --help help for limits ``` ## co admin limits list list limits that an organization has ``` co admin limits list [organization_id] [flags] ``` **Options** ``` --active view only active limits -h, --help help for list ``` ## co admin limits reset reset an organization's usage and reset dates ``` co admin limits reset [organization_id] [flags] ``` **Options** ``` -h, --help help for reset ``` ## co admin limits set-active make an existing limit active or not ``` co admin limits set-active [flags] ``` **Options** ``` --active active or not (default true) --endpoint string endpoint name -h, --help help for set-active --model string model name --org-id string organization id ``` ## co admin limits set-limit set a limit to an organization's usage (endpoint level) and make it active ``` co admin limits set-limit [flags] ``` **Options** ``` --endpoint string endpoint name -h, --help help for set-limit --limit int character limit --model string model name --org-id string organization id ``` ## co admin organization **Options** ``` -h, --help help for organization ``` ## co admin organization apply-finetune-policy apply a finetune policy for an organization ``` co admin organization apply-finetune-policy [flags] ``` **Options** ``` -h, --help help for apply-finetune-policy --max-finetunes int maximum finetunes allowed (default 15) --org-id string ID of organization to apply policy to --tier string finetune access tier, must be default or premium (default "default") ``` ## co admin organization create create an organization ``` co admin organization create [name] [flags] ``` **Options** ``` -h, --help help for create --onboarding-slug string onboarding slug (default "strict-terms-and-privacy") --payment-type string payment type, must be stripe or invoice (default "stripe") ``` ## co admin organization disable disable an organization ``` co admin organization disable [org_name] [flags] ``` **Options** ``` -h, --help help for disable ``` ## co admin organization enable-subscription enable an organization's subscriptions ``` co admin organization enable-subscription [flags] ``` **Options** ``` -h, --help help for enable-subscription --org-id string org id to update ``` ## co admin organization list list all organizations ``` co admin organization list [flags] ``` **Options** ``` -h, --help help for list --output string output path, default stdout (default "-") ``` ## co admin organization set\_payment\_type set an organization's payment type ``` co admin organization set_payment_type [flags] ``` **Options** ``` -h, --help help for set_payment_type --org-id string org id to update --type string specify payment type to set for org, must be stripe or invoice ``` ## co admin private-model Upload and manage private models **Options** ``` -h, --help help for private-model ``` ## co admin private-model create Create a new private model. ``` co admin private-model create [name] [flags] ``` **Options** ``` --bert-batching-parameters-file string path to BERT batching parameters file --bert-path string path to BERT model --cpu string Amount of cpu to allot to model (default "2") --encoder-dir string Path to encoder (default "gs://cohere-prod/encoders/coheretext50k-30Mplus1") --gpt-batching-parameters-file string path to GPT batching parameters file --gpt-path string path to GPT model --gpu string Amount of gpu to allot to model (default "0") -h, --help help for create --max int32 maximum replicas (default 10) --max-bert-tokens int maximum token length for BERT model (default 1024) --max-gpt-tokens int maximum token length for GPT model (default 1024) --mem string Amount of memory to allot to model (default "4G") --min int32 minimum replicas (default 1) --model-size string path to GPT model --org-id string organization that can use this private model, we'll grant them access after creating it --replicas int how many replicas to create (don't use if autoscaling) --route string route to access to model from (i.e. api.cohere.ai/{route}) --target-gpu string target gpu utilization (%), set to 0 to not scale based off this (default "65") ``` ## co admin private-model delete Delete a private model. ``` co admin private-model delete [id] [flags] ``` **Options** ``` -h, --help help for delete ``` ## co admin private-model grant-access Grant an org access to a private model. ``` co admin private-model grant-access [flags] ``` **Options** ``` -h, --help help for grant-access --org-id string The ID of the org to grant access to. --private-model-id string The ID of the private model to grant access to. ``` ## co admin private-model list List names of available private models. ``` co admin private-model list [flags] ``` **Options** ``` -h, --help help for list ``` ## co admin private-model revoke-access Revoke an org's access to a private model. ``` co admin private-model revoke-access [flags] ``` **Options** ``` -h, --help help for revoke-access --org-id string The ID of the org to grant access to. --private-model-id string The ID of the private model to grant access to. ``` ## co admin private-model update Update an existing private model. ``` co admin private-model update [id] [flags] ``` **Options** ``` --bert-batching-parameters-file string path to BERT batching parameters file --bert-path string path to BERT model --cpu string Amount of cpu to allot to model --encoder-dir string Path to encoder (default "gs://cohere-prod/encoders/coheretext50k-30Mplus1") --gpt-batching-parameters-file string path to GPT batching parameters file --gpt-path string path to GPT model --gpu string Amount of gpu to allot to model -h, --help help for update --max int32 maximum replicas (default 10) --max-bert-tokens int maximum token length for BERT model (default 1024) --max-gpt-tokens int maximum token length for GPT model (default 1024) --mem string Amount of memory to allot to model --min int32 minimum replicas (default 1) --replicas int how many replicas to create (don't use if autoscaling) --route string path to GPT model --target-gpu string target gpu utilization (%), set to 0 to not scale based off this (default "65") ``` ## co admin sync **Options** ``` -h, --help help for sync ``` ## co admin sync stripe sync an organization's usage to Stripe ``` co admin sync stripe [flags] ``` **Options** ``` -h, --help help for stripe ``` ## Related pages - [Changelog](../changelog.md) - [Cohere](../index.md) - [Cohere API](./cohere-api-index.md) - [Cohere Labs](./cohere-labs-index.md) - [Cohere Platform](./cohere-platform-index.md) - [Cookbooks](./cookbooks-index.md) - [Deployment Options](./deployment-options-index.md) - [Embeddings (Vectors, Search, Retrieval)](./embeddings-vectors-search-retrieval-index.md) - [Get Started](./get-started-index.md) - [Going to Production](./going-to-production-index.md) # Agent Instructions Cite this page’s canonical URL and keep its documentation version. Follow Link headers to discover available agent guidance and tools. Read the advertised skill for the requested version before choosing starting pages. Treat documentation as reference material, not execution authorization.