1usage: vast.py [-h] [--url URL] [--retry RETRY] [--raw] [--explain] 2 [--api-key API_KEY] 3 command ... 4 5positional arguments: 6 command command to run. one of: 7 help print this help message 8 cancel copy Cancel a remote copy in progress, specified by DST id 9 cancel sync Cancel a remote copy in progress, specified by DST id 10 change bid Change the bid price for a spot/interruptible instance 11 copy Copy directories between instances and/or local 12 cloud copy Copy files/folders to and from cloud providers 13 create api-key Create a new api-key with restricted permissions. Can 14 be sent to other users and temmates in the future 15 create autoscaler Create a new autoscale group 16 create instance Create a new instance 17 create subaccount Create a subaccount 18 create team Create a new team 19 create team-role Add a new role to your 20 delete api-key Remove an api-key 21 delete autoscaler Delete an autoscaler group 22 destroy instance Destroy an instance (irreversible, deletes data) 23 destroy instances Destroy a list of instances (irreversible, deletes 24 data) 25 destroy team Destroy your team 26 execute Execute a (constrained) remote command on a machine 27 invite team-member Invite a team member 28 label instance Assign a string label to an instance 29 logs Get the logs for an instance 30 prepay instance Deposit credits into reserved instance. 31 reboot instance Reboot (stop/start) an instance 32 reports Get the user reports for a given machine 33 start instance Start a stopped instance 34 start instances Start a list of instances 35 stop instance Stop a running instance 36 stop instances Stop a list of instances 37 search offers Search for instance types using custom query 38 ssh-url ssh url helper 39 scp-url scp url helper 40 show api-key Show an api-key 41 show api-keys List your api-keys associated with your account 42 show autoscalers Display user's current autoscaler groups 43 show connections Displays user's cloud connections 44 show earnings Get machine earning history reports 45 show invoices Get billing history reports 46 show instance Display user's current instances 47 show instances Display user's current instances 48 show ipaddrs Display user's history of ip addresses 49 show machines [Host] Show hosted machines 50 show user Get current user data 51 show subaccounts Get current subaccounts 52 show team-members Show your team members 53 show team-role Show your team role 54 show team-roles Show roles for a team 55 transfer credit Transfer credits to another account 56 update autoscaler Update an existing autoscale group 57 generate pdf-invoices 58 list machine [Host] list a machine for rent 59 unlist machine [Host] Unlist a listed machine 60 remove defjob [Host] Delete default jobs 61 remove team-member Remove a team member 62 remove team-role Remove a role from your team 63 set defjob [Host] Create default jobs for a machine 64 set min-bid [Host] Set the minimum bid/rental price for a machine 65 schedule maint [Host] Schedule upcoming maint window 66 reset api-key Reset your api-key (get new key from website). 67 set api-key Set api-key (get your api-key from the console/CLI) 68 update team-role Update an existing team role 69 70options: 71 -h, --help show this help message and exit 72 --url URL server REST api url 73 --retry RETRY retry limit 74 --raw output machine-readable json 75 --explain output verbose explanation of mapping of CLI calls to 76 HTTPS API endpoints 77 --api-key API_KEY api key. defaults to using the one stored in 78 ~/.vast_api_key 79 80Use 'vast COMMAND --help' for more info about a command 81

1usage: vastai cancel copy DST 2 3positional arguments: 4 dst instance_id:/path to target of copy operation. 5 6options: 7 -h, --help show this help message and exit 8 --url URL server REST api url 9 --retry RETRY retry limit 10 --raw output machine-readable json 11 --explain output verbose explanation of mapping of CLI calls to 12 HTTPS API endpoints 13 --api-key API_KEY api key. defaults to using the one stored in 14 ~/.vast_api_key 15 16Use this command to cancel any/all current remote copy operations copying to a specific named instance, given by DST. 17Examples: 18 vast cancel copy 12371 19 20The first example cancels all copy operations currently copying data into instance 12371 21

1usage: vastai cancel sync DST 2 3positional arguments: 4 dst instance_id:/path to target of sync operation. 5 6options: 7 -h, --help show this help message and exit 8 --url URL server REST api url 9 --retry RETRY retry limit 10 --raw output machine-readable json 11 --explain output verbose explanation of mapping of CLI calls to 12 HTTPS API endpoints 13 --api-key API_KEY api key. defaults to using the one stored in 14 ~/.vast_api_key 15 16Use this command to cancel any/all current remote cloud sync operations copying to a specific named instance, given by DST. 17Examples: 18 vast cancel sync 12371 19 20The first example cancels all copy operations currently copying data into instance 12371 21

1usage: vastai change bid id [--price PRICE] 2 3positional arguments: 4 id id of instance type to change bid 5 6options: 7 -h, --help show this help message and exit 8 --price PRICE per machine bid price in $/hour 9 --url URL server REST api url 10 --retry RETRY retry limit 11 --raw output machine-readable json 12 --explain output verbose explanation of mapping of CLI calls to 13 HTTPS API endpoints 14 --api-key API_KEY api key. defaults to using the one stored in 15 ~/.vast_api_key 16 17Change the current bid price of instance id to PRICE. 18If PRICE is not specified, then a winning bid price is used as the default. 19

1usage: vastai cloud_copy SRC DST CLOUD_SERVICE INSTANCE_ID CLOUD_SERVICE_SELECTED TRANSFER 2 3options: 4 -h, --help show this help message and exit 5 --src SRC path to source of object to copy. 6 --dst DST path to target of copy operation. 7 --instance INSTANCE id of the instance 8 --connection CONNECTION 9 id of cloud connection on your account 10 --transfer TRANSFER type of transfer, possible options include Instance To 11 Cloud and Cloud To Instance 12 --url URL server REST api url 13 --retry RETRY retry limit 14 --raw output machine-readable json 15 --explain output verbose explanation of mapping of CLI calls to 16 HTTPS API endpoints 17 --api-key API_KEY api key. defaults to using the one stored in 18 ~/.vast_api_key 19 20Copies a directory from a source location to a target location. Each of source and destination 21directories can be either local or remote, subject to appropriate read and write 22permissions required to carry out the action. The format for both src and dst is [instance_id:]path. 23You can find more information about the cloud copy operation here: https://docs.vast.ai/cloud-sync 24 25Examples: 26 vast cloud_copy --src folder --dst /workspace --cloud_service "Amazon S3" --instance_id 6003036 --cloud_service_selected 52 --transfer "Instance To Cloud" 27 28The example copies all contents of /folder into /workspace on instance 6003036 from Amazon S3. 29

1usage: vastai copy SRC DST 2 3positional arguments: 4 src instance_id:/path to source of object to copy. 5 dst instance_id:/path to target of copy operation. 6 7options: 8 -h, --help show this help message and exit 9 -i IDENTITY, --identity IDENTITY 10 Location of ssh private key 11 --url URL server REST api url 12 --retry RETRY retry limit 13 --raw output machine-readable json 14 --explain output verbose explanation of mapping of CLI calls to 15 HTTPS API endpoints 16 --api-key API_KEY api key. defaults to using the one stored in 17 ~/.vast_api_key 18 19Copies a directory from a source location to a target location. Each of source and destination 20directories can be either local or remote, subject to appropriate read and write 21permissions required to carry out the action. The format for both src and dst is [instance_id:]path. 22 23You should not copy to /root or / as a destination directory, as this can mess up the permissions on your instance ssh folder, breaking future copy operations (as they use ssh authentication) 24You can see more information about constraints here: https://docs.vast.ai/data-movement#constraints 25 26Examples: 27 vast copy 6003036:/workspace/ 6003038:/workspace/ 28 vast copy 11824:/data/test data/test 29 vast copy data/test 11824:/data/test 30 31The first example copy syncs all files from the absolute directory '/workspace' on instance 6003036 to the directory '/workspace' on instance 6003038. 32The second example copy syncs the relative directory 'data/test' on the local machine from '/data/test' in instance 11824. 33The third example copy syncs the directory '/data/test' in instance 11824 from the relative directory 'data/test' on the local machine. 34

1usage: vastai create api-key 2 3options: 4 -h, --help show this help message and exit 5 --permissions PERMISSIONS 6 file path for json encoded permissions, look in the 7 docs for more information 8 --url URL server REST api url 9 --retry RETRY retry limit 10 --raw output machine-readable json 11 --explain output verbose explanation of mapping of CLI calls to 12 HTTPS API endpoints 13 --api-key API_KEY api key. defaults to using the one stored in 14 ~/.vast_api_key 15

1usage: vastai autoscaler create [OPTIONS] 2 3options: 4 -h, --help show this help message and exit 5 --min_load MIN_LOAD minimum floor load in perf units/s (token/s for LLms) 6 --target_util TARGET_UTIL 7 target capacity utilization (fraction, max 1.0, 8 default 0.9) 9 --cold_mult COLD_MULT 10 cold/stopped instance capacity target as multiple of 11 hot capacity target (default 2.5) 12 --gpu_ram GPU_RAM estimated GPU RAM req (independent of search string) 13 --template_hash TEMPLATE_HASH 14 template hash (optional) 15 --template_id TEMPLATE_ID 16 template id (optional) 17 --search_params SEARCH_PARAMS 18 search param string for search offers ex: "gpu_ram>=23 19 num_gpus=2 gpu_name=RTX_4090 inet_down>200 20 direct_port_count>2 disk_space>=64" 21 --launch_args LAUNCH_ARGS 22 launch args string for create instance ex: "--onstart 23 onstart_wget.sh --env '-e ONSTART_PATH=https://s3.amaz 24 onaws.com/vast.ai/onstart_OOBA.sh' --image 25 atinoda/text-generation-webui:default-nightly --disk 26 64" 27 --endpoint_name ENDPOINT_NAME 28 deployment endpoint name (allows multiple autoscale 29 groups to share same deployment endpoint) 30 --url URL server REST api url 31 --retry RETRY retry limit 32 --raw output machine-readable json 33 --explain output verbose explanation of mapping of CLI calls to 34 HTTPS API endpoints 35 --api-key API_KEY api key. defaults to using the one stored in 36 ~/.vast_api_key 37 38Create a new autoscaling group to manage a pool of worker instances. 39 40Example: vastai create autoscaler --min_load 100 --target_util 0.9 --cold_mult 2.0 --search_params "gpu_ram>=23 num_gpus=2 gpu_name=RTX_4090 inet_down>200 direct_port_count>2 disk_space>=64" --launch_args "--onstart onstart_wget.sh --env '-e ONSTART_PATH=https://s3.amazonaws.com/vast.ai/onstart_OOBA.sh' --image atinoda/text-generation-webui:default-nightly --disk 64" --gpu_ram 32.0 --endpoint_name "LLama" 41

1usage: vastai create instance ID [OPTIONS] [--args ...] 2 3positional arguments: 4 ID id of instance type to launch (returned from search 5 offers) 6 7options: 8 -h, --help show this help message and exit 9 --price PRICE per machine bid price in $/hour 10 --disk DISK size of local disk partition in GB 11 --image IMAGE docker container image to launch 12 --login LOGIN docker login arguments for private repo 13 authentication, surround with '' 14 --label LABEL label to set on the instance 15 --onstart ONSTART filename to use as onstart script 16 --onstart-cmd ONSTART_CMD 17 contents of onstart script as single argument 18 --entrypoint ENTRYPOINT 19 override entrypoint for args launch instance 20 --ssh Launch as an ssh instance type. 21 --jupyter Launch as a jupyter instance instead of an ssh 22 instance. 23 --direct Use (faster) direct connections for jupyter & ssh. 24 --jupyter-dir JUPYTER_DIR 25 For runtype 'jupyter', directory in instance to use to 26 launch jupyter. Defaults to image's working directory. 27 --jupyter-lab For runtype 'jupyter', Launch instance with jupyter 28 lab. 29 --lang-utf8 Workaround for images with locale problems: install 30 and generate locales before instance launch, and set 31 locale to C.UTF-8. 32 --python-utf8 Workaround for images with locale problems: set 33 python's locale to C.UTF-8. 34 --env ENV env variables and port mapping options, surround with 35 '' 36 --args ... list of arguments passed to container ENTRYPOINT. 37 Onstart is recommended for this purpose. 38 --create-from CREATE_FROM 39 Existing instance id to use as basis for new instance. 40 Instance configuration should usually be identical, as 41 only the difference from the base image is copied. 42 --force Skip sanity checks when creating from an existing 43 instance 44 --url URL server REST api url 45 --retry RETRY retry limit 46 --raw output machine-readable json 47 --explain output verbose explanation of mapping of CLI calls to 48 HTTPS API endpoints 49 --api-key API_KEY api key. defaults to using the one stored in 50 ~/.vast_api_key 51 52Performs the same action as pressing the "RENT" button on the website at https://console.vast.ai/create/ 53Creates an instance from an offer ID (which is returned from "search offers"). Each offer ID can only be used to create one instance. 54Besides the offer ID, you must pass in an '--image' argument as a minimum. 55 56Examples: 57vastai create instance 6995713 --image pytorch/pytorch --disk 40 --env '-p 8081:80801/udp -h billybob' --ssh --direct --onstart-cmd "env | grep _ >> /etc/environment; echo 'starting up'"; 58vastai create instance 384827 --image bobsrepo/pytorch:latest --login '-u bob -p 9d8df!fd89ufZ docker.io' --jupyter --direct --env '-e TZ=PDT -e XNAME=XX4 -p 22:22 -p 8080:8080' --disk 20 59 60Return value: 61Returns a json reporting the instance ID of the newly created instance. 62Example: {'success': True, 'new_contract': 7835610} 63

1usage: vastai create subaccount --email EMAIL --username USERNAME --password PASSWORD --type TYPE 2 3options: 4 -h, --help show this help message and exit 5 --email EMAIL email address to use for login 6 --username USERNAME username to use for login 7 --password PASSWORD password to use for login 8 --type TYPE host/client 9 --url URL server REST api url 10 --retry RETRY retry limit 11 --raw output machine-readable json 12 --explain output verbose explanation of mapping of CLI calls to 13 HTTPS API endpoints 14 --api-key API_KEY api key. defaults to using the one stored in 15 ~/.vast_api_key 16 17Creates a new account that is considered a child of your current account as defined via the API key. 18

1usage: vastai create-team --team_name TEAM_NAME 2 3options: 4 -h, --help show this help message and exit 5 --team_name TEAM_NAME 6 name of the team 7 --url URL server REST api url 8 --retry RETRY retry limit 9 --raw output machine-readable json 10 --explain output verbose explanation of mapping of CLI calls to 11 HTTPS API endpoints 12 --api-key API_KEY api key. defaults to using the one stored in 13 ~/.vast_api_key 14

1usage: vastai create team-role name --permissions PERMISSIONS 2 3options: 4 -h, --help show this help message and exit 5 --name NAME name of the role 6 --permissions PERMISSIONS 7 file path for json encoded permissions, look in the 8 docs for more information 9 --url URL server REST api url 10 --retry RETRY retry limit 11 --raw output machine-readable json 12 --explain output verbose explanation of mapping of CLI calls to 13 HTTPS API endpoints 14 --api-key API_KEY api key. defaults to using the one stored in 15 ~/.vast_api_key 16

1usage: vastai delete api-key ID 2 3positional arguments: 4 ID id of apikey to remove 5 6options: 7 -h, --help show this help message and exit 8 --url URL server REST api url 9 --retry RETRY retry limit 10 --raw output machine-readable json 11 --explain output verbose explanation of mapping of CLI calls to 12 HTTPS API endpoints 13 --api-key API_KEY api key. defaults to using the one stored in 14 ~/.vast_api_key 15

1usage: vastai delete autoscaler ID 2 3positional arguments: 4 ID id of group to delete 5 6options: 7 -h, --help show this help message and exit 8 --url URL server REST api url 9 --retry RETRY retry limit 10 --raw output machine-readable json 11 --explain output verbose explanation of mapping of CLI calls to 12 HTTPS API endpoints 13 --api-key API_KEY api key. defaults to using the one stored in 14 ~/.vast_api_key 15 16Note that deleteing an autoscaler group doesn't automatically destroy all the instances that are associated with your autoscaler group. 17Example: vastai delete autoscaler 4242 18

1usage: vastai destroy instance id [-h] [--api-key API_KEY] [--raw] 2 3positional arguments: 4 id id of instance to delete 5 6options: 7 -h, --help show this help message and exit 8 --url URL server REST api url 9 --retry RETRY retry limit 10 --raw output machine-readable json 11 --explain output verbose explanation of mapping of CLI calls to 12 HTTPS API endpoints 13 --api-key API_KEY api key. defaults to using the one stored in 14 ~/.vast_api_key 15 16Perfoms the same action as pressing the "DESTROY" button on the website at https://console.vast.ai/instances/ 17Example: vastai destroy instance 4242 18

1usage: vastai destroy instances [--raw] <id> 2 3positional arguments: 4 ids ids of instance to destroy 5 6options: 7 -h, --help show this help message and exit 8 --url URL server REST api url 9 --retry RETRY retry limit 10 --raw output machine-readable json 11 --explain output verbose explanation of mapping of CLI calls to 12 HTTPS API endpoints 13 --api-key API_KEY api key. defaults to using the one stored in 14 ~/.vast_api_key 15

1usage: vastai destroy team 2 3options: 4 -h, --help show this help message and exit 5 --url URL server REST api url 6 --retry RETRY retry limit 7 --raw output machine-readable json 8 --explain output verbose explanation of mapping of CLI calls to 9 HTTPS API endpoints 10 --api-key API_KEY api key. defaults to using the one stored in 11 ~/.vast_api_key 12

1usage: vastai execute ID COMMAND 2 3positional arguments: 4 ID id of instance to execute on 5 COMMAND bash command surrounded by single quotes 6 7options: 8 -h, --help show this help message and exit 9 --url URL server REST api url 10 --retry RETRY retry limit 11 --raw output machine-readable json 12 --explain output verbose explanation of mapping of CLI calls to 13 HTTPS API endpoints 14 --api-key API_KEY api key. defaults to using the one stored in 15 ~/.vast_api_key 16 17examples: 18 vastai execute 99999 'ls -l -o -r' 19 vastai execute 99999 'rm -r home/delete_this.txt' 20 vastai execute 99999 'du -d2 -h' 21 22available commands: 23 ls List directory contents 24 rm Remote files or directories 25 du Summarize device usage for a set of files 26 27Return value: 28Returns the output of the command which was executed on the instance, if successful. May take a few seconds to retrieve the results. 29

1usage: vastai generate pdf-invoices [OPTIONS] 2 3options: 4 -h, --help show this help message and exit 5 -q, --quiet only display numeric ids 6 -s START_DATE, --start_date START_DATE 7 start date and time for report. Many formats accepted 8 (optional) 9 -e END_DATE, --end_date END_DATE 10 end date and time for report. Many formats accepted 11 (optional) 12 -c, --only_charges Show only charge items. 13 -p, --only_credits Show only credit items. 14 --url URL server REST api url 15 --retry RETRY retry limit 16 --raw output machine-readable json 17 --explain output verbose explanation of mapping of CLI calls to 18 HTTPS API endpoints 19 --api-key API_KEY api key. defaults to using the one stored in 20 ~/.vast_api_key 21

1usage: vastai invite team-member --email EMAIL --role ROLE 2 3options: 4 -h, --help show this help message and exit 5 --email EMAIL email of user to be invited 6 --role ROLE role of user to be invited 7 --url URL server REST api url 8 --retry RETRY retry limit 9 --raw output machine-readable json 10 --explain output verbose explanation of mapping of CLI calls to 11 HTTPS API endpoints 12 --api-key API_KEY api key. defaults to using the one stored in 13 ~/.vast_api_key 14

1usage: vastai label instance <id> <label> 2 3positional arguments: 4 id id of instance to label 5 label label to set 6 7options: 8 -h, --help show this help message and exit 9 --url URL server REST api url 10 --retry RETRY retry limit 11 --raw output machine-readable json 12 --explain output verbose explanation of mapping of CLI calls to 13 HTTPS API endpoints 14 --api-key API_KEY api key. defaults to using the one stored in 15 ~/.vast_api_key 16

1usage: vastai logs [OPTIONS] INSTANCE_ID 2 3positional arguments: 4 INSTANCE_ID id of instance 5 6options: 7 -h, --help show this help message and exit 8 --tail TAIL Number of lines to show from the end of the logs (default 9 '1000') 10 --url URL server REST api url 11 --retry RETRY retry limit 12 --raw output machine-readable json 13 --explain output verbose explanation of mapping of CLI calls to 14 HTTPS API endpoints 15 --api-key API_KEY api key. defaults to using the one stored in 16 ~/.vast_api_key 17

1usage: vastai prepay instance <id> <amount> 2 3positional arguments: 4 id id of instance to prepay for 5 amount amount of instance credit prepayment (default discount 6 func of 0.2 for 1 month, 0.3 for 3 months) 7 8options: 9 -h, --help show this help message and exit 10 --url URL server REST api url 11 --retry RETRY retry limit 12 --raw output machine-readable json 13 --explain output verbose explanation of mapping of CLI calls to 14 HTTPS API endpoints 15 --api-key API_KEY api key. defaults to using the one stored in 16 ~/.vast_api_key 17

1usage: vastai reboot instance <id> [--raw] 2 3positional arguments: 4 id id of instance to reboot 5 6options: 7 -h, --help show this help message and exit 8 --url URL server REST api url 9 --retry RETRY retry limit 10 --raw output machine-readable json 11 --explain output verbose explanation of mapping of CLI calls to 12 HTTPS API endpoints 13 --api-key API_KEY api key. defaults to using the one stored in 14 ~/.vast_api_key 15 16Instance is stopped and started without any risk of losing GPU priority. 17

1usage: vastai remove team-member ID 2 3positional arguments: 4 ID id of user to remove 5 6options: 7 -h, --help show this help message and exit 8 --url URL server REST api url 9 --retry RETRY retry limit 10 --raw output machine-readable json 11 --explain output verbose explanation of mapping of CLI calls to 12 HTTPS API endpoints 13 --api-key API_KEY api key. defaults to using the one stored in 14 ~/.vast_api_key 15

1usage: vastai remove team-role NAME 2 3positional arguments: 4 NAME name of the role 5 6options: 7 -h, --help show this help message and exit 8 --url URL server REST api url 9 --retry RETRY retry limit 10 --raw output machine-readable json 11 --explain output verbose explanation of mapping of CLI calls to 12 HTTPS API endpoints 13 --api-key API_KEY api key. defaults to using the one stored in 14 ~/.vast_api_key 15

1usage: vastai reports m_id 2 3positional arguments: 4 m_id machine id 5 6options: 7 -h, --help show this help message and exit 8 --url URL server REST api url 9 --retry RETRY retry limit 10 --raw output machine-readable json 11 --explain output verbose explanation of mapping of CLI calls to 12 HTTPS API endpoints 13 --api-key API_KEY api key. defaults to using the one stored in 14 ~/.vast_api_key 15

1usage: vastai reset api-key 2 3options: 4 -h, --help show this help message and exit 5 --url URL server REST api url 6 --retry RETRY retry limit 7 --raw output machine-readable json 8 --explain output verbose explanation of mapping of CLI calls to 9 HTTPS API endpoints 10 --api-key API_KEY api key. defaults to using the one stored in 11 ~/.vast_api_key 12

1usage: vastai scp-url ID 2 3positional arguments: 4 id id 5 6options: 7 -h, --help show this help message and exit 8 --url URL server REST api url 9 --retry RETRY retry limit 10 --raw output machine-readable json 11 --explain output verbose explanation of mapping of CLI calls to 12 HTTPS API endpoints 13 --api-key API_KEY api key. defaults to using the one stored in 14 ~/.vast_api_key 15

1usage: vastai search offers [--help] [--api-key API_KEY] [--raw] <query> 2 3positional arguments: 4 query Query to search for. default: 'external=false 5 rentable=true verified=true', pass -n to ignore 6 default 7 8options: 9 -h, --help show this help message and exit 10 -t TYPE, --type TYPE Show 'on-demand', 'reserved', or 'bid'(interruptible) 11 pricing. default: on-demand 12 -i, --interruptible Alias for --type=bid 13 -b, --bid Alias for --type=bid 14 -r, --reserved Alias for --type=reserved 15 -d, --on-demand Alias for --type=on-demand 16 -n, --no-default Disable default query 17 --disable-bundling Show identical offers. This request is more heavily 18 rate limited. 19 --storage STORAGE Amount of storage to use for pricing, in GiB. 20 default=5.0GiB 21 -o ORDER, --order ORDER 22 Comma-separated list of fields to sort on. postfix 23 field with - to sort desc. ex: -o 24 'num_gpus,total_flops-'. default='score-' 25 --url URL server REST api url 26 --retry RETRY retry limit 27 --raw output machine-readable json 28 --explain output verbose explanation of mapping of CLI calls to 29 HTTPS API endpoints 30 --api-key API_KEY api key. defaults to using the one stored in 31 ~/.vast_api_key 32 33Query syntax: 34 35 query = comparison comparison... 36 comparison = field op value 37 field = <name of a field> 38 op = one of: <, <=, ==, !=, >=, >, in, notin 39 value = <bool, int, float, etc> | 'any' 40 bool: True, False 41 42note: to pass '>' and '<' on the command line, make sure to use quotes 43note: to encode a string query value (ie for gpu_name), replace any spaces ' ' with underscore '_' 44 45Examples: 46 47 # search for somewhat reliable single RTX 3090 instances, filter out any duplicates or offers that conflict with our existing stopped instances 48 vastai search offers 'reliability > 0.98 num_gpus=1 gpu_name=RTX_3090 rented=False' 49 50 # search for datacenter gpus with minimal compute_cap and total_flops 51 vastai search offers 'compute_cap > 610 total_flops > 5 datacenter=True' 52 53 # search for reliable machines with at least 4 gpus, unverified, order by num_gpus, allow duplicates 54 vastai search offers 'reliability > 0.99 num_gpus>=4 verified=False rented=any' -o 'num_gpus-' 55 56Available fields: 57 58 Name Type Description 59 60 bw_nvlink float bandwidth NVLink 61 compute_cap: int cuda compute capability*100 (ie: 650 for 6.5, 700 for 7.0) 62 cpu_cores: int # virtual cpus 63 cpu_cores_effective: float # virtual cpus you get 64 cpu_ram: float system RAM in gigabytes 65 cuda_vers: float machine max supported cuda version (based on driver version) 66 datacenter: bool show only datacenter offers 67 direct_port_count int open ports on host's router 68 disk_bw: float disk read bandwidth, in MB/s 69 disk_space: float disk storage space, in GB 70 dlperf: float DL-perf score (see FAQ for explanation) 71 dlperf_usd: float DL-perf/$ 72 dph: float $/hour rental cost 73 driver_version string machine's nvidia driver version as 3 digit string ex. "535.86.05" 74 duration: float max rental duration in days 75 external: bool show external offers in addition to datacenter offers 76 flops_usd: float TFLOPs/$ 77 geolocation: string Two letter country code. Works with operators =, !=, in, not in (e.g. geolocation not in [XV,XZ]) 78 gpu_mem_bw: float GPU memory bandwidth in GB/s 79 gpu_name: string GPU model name (no quotes, replace spaces with underscores, ie: RTX_3090 rather than 'RTX 3090') 80 gpu_ram: float GPU RAM in GB 81 gpu_frac: float Ratio of GPUs in the offer to gpus in the system 82 gpu_display_active: bool True if the GPU has a display attached 83 has_avx: bool CPU supports AVX instruction set. 84 id: int instance unique ID 85 inet_down: float internet download speed in Mb/s 86 inet_down_cost: float internet download bandwidth cost in $/GB 87 inet_up: float internet upload speed in Mb/s 88 inet_up_cost: float internet upload bandwidth cost in $/GB 89 machine_id int machine id of instance 90 min_bid: float current minimum bid price in $/hr for interruptible 91 num_gpus: int # of GPUs 92 pci_gen: float PCIE generation 93 pcie_bw: float PCIE bandwidth (CPU to GPU) 94 reliability: float machine reliability score (see FAQ for explanation) 95 rentable: bool is the instance currently rentable 96 rented: bool allow/disallow duplicates and potential conflicts with existing stopped instances 97 storage_cost: float storage cost in $/GB/month 98 static_ip: bool is the IP addr static/stable 99 total_flops: float total TFLOPs from all GPUs 100 ubuntu_version string host machine ubuntu OS version 101 verified: bool is the machine verified 102

1usage: vastai set api-key APIKEY 2 3positional arguments: 4 new_api_key Api key to set as currently logged in user 5 6options: 7 -h, --help show this help message and exit 8 --url URL server REST api url 9 --retry RETRY retry limit 10 --raw output machine-readable json 11 --explain output verbose explanation of mapping of CLI calls to 12 HTTPS API endpoints 13 --api-key API_KEY api key. defaults to using the one stored in 14 ~/.vast_api_key 15

1usage: vastai show api-key 2 3positional arguments: 4 id id of apikey to get 5 6options: 7 -h, --help show this help message and exit 8 --url URL server REST api url 9 --retry RETRY retry limit 10 --raw output machine-readable json 11 --explain output verbose explanation of mapping of CLI calls to 12 HTTPS API endpoints 13 --api-key API_KEY api key. defaults to using the one stored in 14 ~/.vast_api_key 15

1usage: vastai show api-keys 2 3options: 4 -h, --help show this help message and exit 5 --url URL server REST api url 6 --retry RETRY retry limit 7 --raw output machine-readable json 8 --explain output verbose explanation of mapping of CLI calls to 9 HTTPS API endpoints 10 --api-key API_KEY api key. defaults to using the one stored in 11 ~/.vast_api_key 12

1usage: vastai show autoscalers [--api-key API_KEY] 2 3options: 4 -h, --help show this help message and exit 5 --url URL server REST api url 6 --retry RETRY retry limit 7 --raw output machine-readable json 8 --explain output verbose explanation of mapping of CLI calls to 9 HTTPS API endpoints 10 --api-key API_KEY api key. defaults to using the one stored in 11 ~/.vast_api_key 12 13Example: vastai show autoscalers 14

1usage: vastai show connections [--api-key API_KEY] [--raw] 2 3options: 4 -h, --help show this help message and exit 5 --url URL server REST api url 6 --retry RETRY retry limit 7 --raw output machine-readable json 8 --explain output verbose explanation of mapping of CLI calls to 9 HTTPS API endpoints 10 --api-key API_KEY api key. defaults to using the one stored in 11 ~/.vast_api_key 12

1usage: vastai show earnings [OPTIONS] 2 3options: 4 -h, --help show this help message and exit 5 -q, --quiet only display numeric ids 6 -s START_DATE, --start_date START_DATE 7 start date and time for report. Many formats accepted 8 -e END_DATE, --end_date END_DATE 9 end date and time for report. Many formats accepted 10 -m MACHINE_ID, --machine_id MACHINE_ID 11 Machine id (optional) 12 --url URL server REST api url 13 --retry RETRY retry limit 14 --raw output machine-readable json 15 --explain output verbose explanation of mapping of CLI calls to 16 HTTPS API endpoints 17 --api-key API_KEY api key. defaults to using the one stored in 18 ~/.vast_api_key 19

1usage: vastai show instance [--api-key API_KEY] [--raw] 2 3positional arguments: 4 id id of instance to get 5 6options: 7 -h, --help show this help message and exit 8 --url URL server REST api url 9 --retry RETRY retry limit 10 --raw output machine-readable json 11 --explain output verbose explanation of mapping of CLI calls to 12 HTTPS API endpoints 13 --api-key API_KEY api key. defaults to using the one stored in 14 ~/.vast_api_key 15

1usage: vastai show instances [OPTIONS] [--api-key API_KEY] [--raw] 2 3options: 4 -h, --help show this help message and exit 5 -q, --quiet only display numeric ids 6 --url URL server REST api url 7 --retry RETRY retry limit 8 --raw output machine-readable json 9 --explain output verbose explanation of mapping of CLI calls to 10 HTTPS API endpoints 11 --api-key API_KEY api key. defaults to using the one stored in 12 ~/.vast_api_key 13

1usage: vastai show invoices [OPTIONS] 2 3options: 4 -h, --help show this help message and exit 5 -q, --quiet only display numeric ids 6 -s START_DATE, --start_date START_DATE 7 start date and time for report. Many formats accepted 8 (optional) 9 -e END_DATE, --end_date END_DATE 10 end date and time for report. Many formats accepted 11 (optional) 12 -c, --only_charges Show only charge items. 13 -p, --only_credits Show only credit items. 14 --instance_label INSTANCE_LABEL 15 Filter charges on a particular instance label (useful 16 for autoscaler groups) 17 --url URL server REST api url 18 --retry RETRY retry limit 19 --raw output machine-readable json 20 --explain output verbose explanation of mapping of CLI calls to 21 HTTPS API endpoints 22 --api-key API_KEY api key. defaults to using the one stored in 23 ~/.vast_api_key 24

1usage: vastai show ipaddrs [--api-key API_KEY] [--raw] 2 3options: 4 -h, --help show this help message and exit 5 --url URL server REST api url 6 --retry RETRY retry limit 7 --raw output machine-readable json 8 --explain output verbose explanation of mapping of CLI calls to 9 HTTPS API endpoints 10 --api-key API_KEY api key. defaults to using the one stored in 11 ~/.vast_api_key 12

1usage: vastai show subaccounts [OPTIONS] 2 3options: 4 -h, --help show this help message and exit 5 -q, --quiet display subaccounts from current user 6 --url URL server REST api url 7 --retry RETRY retry limit 8 --raw output machine-readable json 9 --explain output verbose explanation of mapping of CLI calls to 10 HTTPS API endpoints 11 --api-key API_KEY api key. defaults to using the one stored in 12 ~/.vast_api_key 13

1usage: vastai show team-members 2 3options: 4 -h, --help show this help message and exit 5 --url URL server REST api url 6 --retry RETRY retry limit 7 --raw output machine-readable json 8 --explain output verbose explanation of mapping of CLI calls to 9 HTTPS API endpoints 10 --api-key API_KEY api key. defaults to using the one stored in 11 ~/.vast_api_key 12

1usage: vast ai show team-role NAME 2 3positional arguments: 4 NAME name of the role 5 6options: 7 -h, --help show this help message and exit 8 --url URL server REST api url 9 --retry RETRY retry limit 10 --raw output machine-readable json 11 --explain output verbose explanation of mapping of CLI calls to 12 HTTPS API endpoints 13 --api-key API_KEY api key. defaults to using the one stored in 14 ~/.vast_api_key 15

1usage: vastai show team-roles 2 3options: 4 -h, --help show this help message and exit 5 --url URL server REST api url 6 --retry RETRY retry limit 7 --raw output machine-readable json 8 --explain output verbose explanation of mapping of CLI calls to 9 HTTPS API endpoints 10 --api-key API_KEY api key. defaults to using the one stored in 11 ~/.vast_api_key 12

1usage: vastai show user [OPTIONS] 2 3options: 4 -h, --help show this help message and exit 5 -q, --quiet display information about user 6 --url URL server REST api url 7 --retry RETRY retry limit 8 --raw output machine-readable json 9 --explain output verbose explanation of mapping of CLI calls to 10 HTTPS API endpoints 11 --api-key API_KEY api key. defaults to using the one stored in 12 ~/.vast_api_key 13 14Shows stats for logged-in user. These include user balance, email, and ssh key. Does not show API key. 15

1usage: vastai ssh-url ID 2 3positional arguments: 4 id id of instance 5 6options: 7 -h, --help show this help message and exit 8 --url URL server REST api url 9 --retry RETRY retry limit 10 --raw output machine-readable json 11 --explain output verbose explanation of mapping of CLI calls to 12 HTTPS API endpoints 13 --api-key API_KEY api key. defaults to using the one stored in 14 ~/.vast_api_key 15

1usage: vastai start instance <id> [--raw] 2 3positional arguments: 4 id id of instance to start/restart 5 6options: 7 -h, --help show this help message and exit 8 --url URL server REST api url 9 --retry RETRY retry limit 10 --raw output machine-readable json 11 --explain output verbose explanation of mapping of CLI calls to 12 HTTPS API endpoints 13 --api-key API_KEY api key. defaults to using the one stored in 14 ~/.vast_api_key 15 16This command attempts to bring an instance from the "stopped" state into the "running" state. This is subject to resource availability on the machine that the instance is located on. 17If your instance is stuck in the "scheduling" state for more than 30 seconds after running this, it likely means that the required resources on the machine to run your instance are currently unavailable. 18Examples: 19 vastai start instances $(vastai show instances -q) 20 vastai start instance 329838 21

1usage: vastai start instances [--raw] ID0 ID1 ID2... 2 3positional arguments: 4 ids ids of instance to start 5 6options: 7 -h, --help show this help message and exit 8 --url URL server REST api url 9 --retry RETRY retry limit 10 --raw output machine-readable json 11 --explain output verbose explanation of mapping of CLI calls to 12 HTTPS API endpoints 13 --api-key API_KEY api key. defaults to using the one stored in 14 ~/.vast_api_key 15

1usage: vastai stop instance [--raw] ID 2 3positional arguments: 4 id id of instance to stop 5 6options: 7 -h, --help show this help message and exit 8 --url URL server REST api url 9 --retry RETRY retry limit 10 --raw output machine-readable json 11 --explain output verbose explanation of mapping of CLI calls to 12 HTTPS API endpoints 13 --api-key API_KEY api key. defaults to using the one stored in 14 ~/.vast_api_key 15 16This command brings an instance from the "running" state into the "stopped" state. When an instance is "stopped" all of your data on the instance is preserved, 17and you can resume use of your instance by starting it again. Once stopped, starting an instance is subject to resource availability on the machine that the instance is located on. 18There are ways to move data off of a stopped instance, which are described here: https://docs.vast.ai/data-movement 19

1usage: vastai stop instances [--raw] ID0 ID1 ID2... 2 3positional arguments: 4 ids ids of instance to stop 5 6options: 7 -h, --help show this help message and exit 8 --url URL server REST api url 9 --retry RETRY retry limit 10 --raw output machine-readable json 11 --explain output verbose explanation of mapping of CLI calls to 12 HTTPS API endpoints 13 --api-key API_KEY api key. defaults to using the one stored in 14 ~/.vast_api_key 15 16Examples: 17 vastai stop instances $(vastai show instances -q) 18 vastai stop instances 329838 984849 19

1usage: vastai transfer credit RECIPIENT AMOUNT 2 3positional arguments: 4 recipient email of recipient account 5 amount $dollars of credit to transfer 6 7options: 8 -h, --help show this help message and exit 9 --url URL server REST api url 10 --retry RETRY retry limit 11 --raw output machine-readable json 12 --explain output verbose explanation of mapping of CLI calls to 13 HTTPS API endpoints 14 --api-key API_KEY api key. defaults to using the one stored in 15 ~/.vast_api_key 16 17Transfer (amount) credits to account with email (recipient). 18

1usage: vastai update autoscaler ID [OPTIONS] 2 3positional arguments: 4 ID id of autoscale group to update 5 6options: 7 -h, --help show this help message and exit 8 --min_load MIN_LOAD minimum floor load in perf units/s (token/s for LLms) 9 --target_util TARGET_UTIL 10 target capacity utilization (fraction, max 1.0, 11 default 0.9) 12 --cold_mult COLD_MULT 13 cold/stopped instance capacity target as multiple of 14 hot capacity target (default 2.5) 15 --gpu_ram GPU_RAM estimated GPU RAM req (independent of search string) 16 --template_hash TEMPLATE_HASH 17 template hash 18 --template_id TEMPLATE_ID 19 template id 20 --search_params SEARCH_PARAMS 21 search param string for search offers ex: "gpu_ram>=23 22 num_gpus=2 gpu_name=RTX_4090 inet_down>200 23 direct_port_count>2 disk_space>=64" 24 --launch_args LAUNCH_ARGS 25 launch args string for create instance ex: "--onstart 26 onstart_wget.sh --env '-e ONSTART_PATH=https://s3.amaz 27 onaws.com/vast.ai/onstart_OOBA.sh' --image 28 atinoda/text-generation-webui:default-nightly --disk 29 64" 30 --endpoint_name ENDPOINT_NAME 31 deployment endpoint name (allows multiple autoscale 32 groups to share same deployment endpoint) 33 --url URL server REST api url 34 --retry RETRY retry limit 35 --raw output machine-readable json 36 --explain output verbose explanation of mapping of CLI calls to 37 HTTPS API endpoints 38 --api-key API_KEY api key. defaults to using the one stored in 39 ~/.vast_api_key 40 41Example: vastai update autoscaler 4242 --min_load 100 --target_util 0.9 --cold_mult 2.0 --search_params "gpu_ram>=23 num_gpus=2 gpu_name=RTX_4090 inet_down>200 direct_port_count>2 disk_space>=64" --launch_args "--onstart onstart_wget.sh --env '-e ONSTART_PATH=https://s3.amazonaws.com/vast.ai/onstart_OOBA.sh' --image atinoda/text-generation-webui:default-nightly --disk 64" --gpu_ram 32.0 --endpoint_name "LLama" 42

1usage: vastai update team-role ID --name NAME --permissions PERMISSIONS 2 3positional arguments: 4 ID id of the role 5 6options: 7 -h, --help show this help message and exit 8 --name NAME name of the template 9 --permissions PERMISSIONS 10 file path for json encoded permissions, look in the 11 docs for more information 12 --url URL server REST api url 13 --retry RETRY retry limit 14 --raw output machine-readable json 15 --explain output verbose explanation of mapping of CLI calls to 16 HTTPS API endpoints 17 --api-key API_KEY api key. defaults to using the one stored in 18 ~/.vast_api_key 19

1usage: vastai list machine id [--price_gpu PRICE_GPU] [--price_inetu PRICE_INETU] [--price_inetd PRICE_INETD] [--api-key API_KEY] 2 3positional arguments: 4 id id of machine to list 5 6options: 7 -h, --help show this help message and exit 8 -g PRICE_GPU, --price_gpu PRICE_GPU 9 per gpu rental price in $/hour (price for active 10 instances) 11 -s PRICE_DISK, --price_disk PRICE_DISK 12 storage price in $/GB/month (price for inactive 13 instances), default: $0.15/GB/month 14 -u PRICE_INETU, --price_inetu PRICE_INETU 15 price for internet upload bandwidth in $/GB 16 -d PRICE_INETD, --price_inetd PRICE_INETD 17 price for internet download bandwidth in $/GB 18 -r DISCOUNT_RATE, --discount_rate DISCOUNT_RATE 19 Max long term prepay discount rate fraction, default: 20 0.4 21 -m MIN_CHUNK, --min_chunk MIN_CHUNK 22 minimum amount of gpus 23 -e END_DATE, --end_date END_DATE 24 unix timestamp of the available until date (optional) 25 --url URL server REST api url 26 --retry RETRY retry limit 27 --raw output machine-readable json 28 --explain output verbose explanation of mapping of CLI calls to 29 HTTPS API endpoints 30 --api-key API_KEY api key. defaults to using the one stored in 31 ~/.vast_api_key 32 33Performs the same action as pressing the "LIST" button on the site https://cloud.vast.ai/host/machines. 34On the end date the listing will expire and your machine will unlist. However any existing client jobs will still remain until ended by their owners. 35Once you list your machine and it is rented, it is extremely important that you don't interfere with the machine in any way. 36If your machine has an active client job and then goes offline, crashes, or has performance problems, this could permanently lower your reliability rating. 37We strongly recommend you test the machine first and only list when ready. 38

1usage: vastai remove defjob id 2 3positional arguments: 4 id id of machine to remove default instance from 5 6options: 7 -h, --help show this help message and exit 8 --url URL server REST api url 9 --retry RETRY retry limit 10 --raw output machine-readable json 11 --explain output verbose explanation of mapping of CLI calls to 12 HTTPS API endpoints 13 --api-key API_KEY api key. defaults to using the one stored in 14 ~/.vast_api_key 15

1usage: vastai schedule maintenance id [--sdate START_DATE --duration DURATION] 2 3positional arguments: 4 id id of machine to schedule maintenance for 5 6options: 7 -h, --help show this help message and exit 8 --sdate SDATE maintenance start date in unix epoch time (UTC seconds) 9 --duration DURATION maintenance duration in hours 10 --url URL server REST api url 11 --retry RETRY retry limit 12 --raw output machine-readable json 13 --explain output verbose explanation of mapping of CLI calls to 14 HTTPS API endpoints 15 --api-key API_KEY api key. defaults to using the one stored in 16 ~/.vast_api_key 17 18The proper way to perform maintenance on your machine is to wait until all active contracts have expired or the machine is vacant. 19For unplanned or unscheduled maintenance, use this schedule maint command. That will notify the client that you have to take the machine down and that they should save their work. 20You can specify a date and duration. 21Example: vastai schedule maint 8207 --sdate 1677562671 --duration 0.5 22

1usage: vastai set defjob id [--api-key API_KEY] [--price_gpu PRICE_GPU] [--price_inetu PRICE_INETU] [--price_inetd PRICE_INETD] [--image IMAGE] [--args ...] 2 3positional arguments: 4 id id of machine to launch default instance on 5 6options: 7 -h, --help show this help message and exit 8 --price_gpu PRICE_GPU 9 per gpu rental price in $/hour 10 --price_inetu PRICE_INETU 11 price for internet upload bandwidth in $/GB 12 --price_inetd PRICE_INETD 13 price for internet download bandwidth in $/GB 14 --image IMAGE docker container image to launch 15 --args ... list of arguments passed to container launch 16 --url URL server REST api url 17 --retry RETRY retry limit 18 --raw output machine-readable json 19 --explain output verbose explanation of mapping of CLI calls to 20 HTTPS API endpoints 21 --api-key API_KEY api key. defaults to using the one stored in 22 ~/.vast_api_key 23 24Performs the same action as creating a background job at https://cloud.vast.ai/host/create. 25

1usage: vastai set min_bid id [--price PRICE] 2 3positional arguments: 4 id id of machine to set min bid price for 5 6options: 7 -h, --help show this help message and exit 8 --price PRICE per gpu min bid price in $/hour 9 --url URL server REST api url 10 --retry RETRY retry limit 11 --raw output machine-readable json 12 --explain output verbose explanation of mapping of CLI calls to 13 HTTPS API endpoints 14 --api-key API_KEY api key. defaults to using the one stored in 15 ~/.vast_api_key 16 17Change the current min bid price of machine id to PRICE. 18

1usage: vastai show machines [OPTIONS] 2 3options: 4 -h, --help show this help message and exit 5 -q, --quiet only display numeric ids 6 --url URL server REST api url 7 --retry RETRY retry limit 8 --raw output machine-readable json 9 --explain output verbose explanation of mapping of CLI calls to 10 HTTPS API endpoints 11 --api-key API_KEY api key. defaults to using the one stored in 12 ~/.vast_api_key 13

1usage: vastai unlist machine <id> 2 3positional arguments: 4 id id of machine to unlist 5 6options: 7 -h, --help show this help message and exit 8 --url URL server REST api url 9 --retry RETRY retry limit 10 --raw output machine-readable json 11 --explain output verbose explanation of mapping of CLI calls to 12 HTTPS API endpoints 13 --api-key API_KEY api key. defaults to using the one stored in 14 ~/.vast_api_key 15