Properties
| Name | Type | Description |
|---|---|---|
| executor_id | string | |
| ports_mapping | object | |
| pod_name | string | |
| ssh_connect_cmd | string | |
| gpu_name | string | |
| gpu_count | string | |
| cpu_name | string | |
| cpu_count | integer | |
| ram_total | integer | |
| status | string | |
| is_favorite | object | |
| updated_at | object | |
| created_at | object | |
| removal_scheduled_at | object | |
| termination_hours | object | |
| jupyter_installation_status | object | |
| jupyter_url | object | |
| enable_jupyter | object | |
| enable_volume_encryption | object | |
| volume_encryption_status | object | |
| local_volume_path | object | |
| price | object | |
| pending_price | object | Hourly USD price this pod moves to at price_change_effective_at: the node's listed price is now above what this pod pays. null when no change is pending. |
| price_change_effective_at | object | UTC time the pending price takes effect. Null when no change is pending or while its notification e-mail is held. Notices are limited to one per pod per 24 hours. |
| price_change_notice_hours | integer | Hours a pending raise waits after the renter is e-mailed about it before it reaches the pod (24 on a Spot node). |
| no_interrupt_until | object | UTC time before which the provider may not start a planned maintenance or removal notice on this pod's node. null when the renter set none. |
| cluster_id | object | |
| cluster_node_index | object | |
| created_by_api_key_id | object | |
| api_key_name | object | |
| workspace_id | object | |
| volume_limit_gb | object | Size in GB of the pod's local /root volume. null on a pod created without a storage limit. |
| spend_to_date | number | USD billed to this rental so far, summed from the billing ledger, not price × elapsed time. The ledger accrues every 5 minutes and settles on deletion, so this lags real time by at most one accrual in |
| billed_until | object | UTC time up to which spend_to_date is settled: the ledger's last accrual tick for this rental. null until the first charge is booked (a pod in its first minutes, or one that never became billable). |
| id | string | |
| template | object | |
| executor | object | |
| volume | object | |
| ssh_keys | object | |
| notice_period | object | |
| removal_at | object | |
| skip_add_agent_ssh_key | object | |
| is_manual_rental | object | |
| estimated_ready_seconds | object | Seconds left of the time-to-ready estimate made when the rent was accepted; 0 once it has run over. Only while the pod is PENDING or REBOOT_PENDING. |
| eta_basis | object | What the estimate was made from: this node's recent creates (executor_history), its GPU class (class_median) or the fleet default (fleet_default); cold_pull_estimate when an image pull the node does n |
| phase | object | The step the create has reached, read from the validator's log lines. |
| rental_tier | object | |
| starting_ssh_connect_cmd | object | The SSH command the pod will answer on, set while it is PENDING or REBOOT_PENDING once its SSH port is known, before the API reports the pod RUNNING. Null while the port is not known yet, and in every |
| create_progress | object | While the pod is PENDING or REBOOT_PENDING: `stage` (the step the create has reached, same text as `phase`), `message` (the validator's latest create log line, null before the first) and `updated_at` |
| node_offline | boolean | The node failed its health checks (not responding) or is marked inactive. The pod keeps its status; it may not be reachable until the node is back. |
| billing_paused | boolean | A billed pod on a node marked inactive or read as down by the validator: each 5-minute charge is $0 until the node is back (see Billing, When a pod is charged). |
| node_outage | object | Set while `node_offline` is true: since when, and how the node's outages in the last 30 days ended. History only, not a prediction. Null otherwise. |
| ssh_unreachable_since | object | Set while the pod is RUNNING but its SSH port stopped answering the validator's probe: since when (ISO-8601 UTC). Null otherwise. |
| pod_status | object | |
| provider_discord_connected | object | |
| last_event | object | |
| machine_changed_at | object | |
| restore_backup_id | object | |
| restore_backup_completed_at | object | |
| disk_used_gb | object | GB of the node's disk the pod uses: its container, including Docker-in-Docker data, and its /root volume, as last measured. null until measured. |
| disk_size_gb | object | GB of disk the pod was given: its /root volume plus its pod disk. null on a pod created without a storage limit. |
| cluster_overlay_ip | object | This node's address on the cluster's private overlay — where its peers reach it. |
JSON Schema
{
"$schema": "https://json-schema.org/draft/2020-12/schema",
"$id": "https://artifacts.apis.io/lium-io/json-schema/lium-io-pod-detail-response-schema.json",
"title": "PodDetailResponse",
"x-generated": "2026-11-17",
"x-method": "derived",
"x-generator": "derive-json-schema.py",
"x-source": "openapi/lium-io-openapi.json#/components/schemas/PodDetailResponse",
"properties": {
"executor_id": {
"type": "string",
"format": "uuid",
"title": "Executor Id"
},
"ports_mapping": {
"anyOf": [
{
"additionalProperties": true,
"type": "object"
},
{
"type": "string"
}
],
"title": "Ports Mapping"
},
"pod_name": {
"type": "string",
"title": "Pod Name"
},
"ssh_connect_cmd": {
"type": "string",
"title": "Ssh Connect Cmd"
},
"gpu_name": {
"type": "string",
"title": "Gpu Name"
},
"gpu_count": {
"type": "string",
"title": "Gpu Count"
},
"cpu_name": {
"type": "string",
"title": "Cpu Name"
},
"cpu_count": {
"type": "integer",
"title": "Cpu Count"
},
"ram_total": {
"type": "integer",
"title": "Ram Total"
},
"status": {
"type": "string",
"title": "Status"
},
"is_favorite": {
"anyOf": [
{
"type": "boolean"
},
{
"type": "null"
}
],
"title": "Is Favorite"
},
"updated_at": {
"anyOf": [
{
"type": "string",
"format": "date-time"
},
{
"type": "null"
}
],
"title": "Updated At"
},
"created_at": {
"anyOf": [
{
"type": "string",
"format": "date-time"
},
{
"type": "null"
}
],
"title": "Created At"
},
"removal_scheduled_at": {
"anyOf": [
{
"type": "string",
"format": "date-time"
},
{
"type": "null"
}
],
"title": "Removal Scheduled At"
},
"termination_hours": {
"anyOf": [
{
"type": "integer"
},
{
"type": "null"
}
],
"title": "Termination Hours"
},
"jupyter_installation_status": {
"anyOf": [
{
"$ref": "#/$defs/JupyterInstallationStatus"
},
{
"type": "null"
}
]
},
"jupyter_url": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"title": "Jupyter Url"
},
"enable_jupyter": {
"anyOf": [
{
"type": "boolean"
},
{
"type": "null"
}
],
"title": "Enable Jupyter"
},
"enable_volume_encryption": {
"anyOf": [
{
"type": "boolean"
},
{
"type": "null"
}
],
"title": "Enable Volume Encryption"
},
"volume_encryption_status": {
"anyOf": [
{
"$ref": "#/$defs/VolumeEncryptionStatus"
},
{
"type": "null"
}
]
},
"local_volume_path": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"title": "Local Volume Path"
},
"price": {
"anyOf": [
{
"type": "number"
},
{
"type": "null"
}
],
"title": "Price"
},
"pending_price": {
"anyOf": [
{
"type": "number"
},
{
"type": "null"
}
],
"title": "Pending Price",
"description": "Hourly USD price this pod moves to at price_change_effective_at: the node's listed price is now above what this pod pays. null when no change is pending."
},
"price_change_effective_at": {
"anyOf": [
{
"type": "string",
"format": "date-time"
},
{
"type": "null"
}
],
"title": "Price Change Effective At",
"description": "UTC time the pending price takes effect. Null when no change is pending or while its notification e-mail is held. Notices are limited to one per pod per 24 hours."
},
"price_change_notice_hours": {
"type": "integer",
"title": "Price Change Notice Hours",
"description": "Hours a pending raise waits after the renter is e-mailed about it before it reaches the pod (24 on a Spot node)."
},
"no_interrupt_until": {
"anyOf": [
{
"type": "string",
"format": "date-time"
},
{
"type": "null"
}
],
"title": "No Interrupt Until",
"description": "UTC time before which the provider may not start a planned maintenance or removal notice on this pod's node. null when the renter set none."
},
"cluster_id": {
"anyOf": [
{
"type": "string",
"format": "uuid"
},
{
"type": "null"
}
],
"title": "Cluster Id"
},
"cluster_node_index": {
"anyOf": [
{
"type": "integer"
},
{
"type": "null"
}
],
"title": "Cluster Node Index"
},
"created_by_api_key_id": {
"anyOf": [
{
"type": "string",
"format": "uuid"
},
{
"type": "null"
}
],
"title": "Created By Api Key Id"
},
"api_key_name": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"title": "Api Key Name"
},
"workspace_id": {
"anyOf": [
{
"type": "string",
"format": "uuid"
},
{
"type": "null"
}
],
"title": "Workspace Id"
},
"volume_limit_gb": {
"anyOf": [
{
"type": "integer"
},
{
"type": "null"
}
],
"title": "Volume Limit Gb",
"description": "Size in GB of the pod's local /root volume. null on a pod created without a storage limit."
},
"spend_to_date": {
"type": "number",
"title": "Spend To Date",
"description": "USD billed to this rental so far, summed from the billing ledger, not price × elapsed time. The ledger accrues every 5 minutes and settles on deletion, so this lags real time by at most one accrual interval; 0.0 until the first charge is booked.",
"default": 0.0
},
"billed_until": {
"anyOf": [
{
"type": "string",
"format": "date-time"
},
{
"type": "null"
}
],
"title": "Billed Until",
"description": "UTC time up to which spend_to_date is settled: the ledger's last accrual tick for this rental. null until the first charge is booked (a pod in its first minutes, or one that never became billable)."
},
"id": {
"type": "string",
"format": "uuid",
"title": "Id"
},
"template": {
"anyOf": [
{
"$ref": "#/$defs/TemplateBaseResponse"
},
{
"type": "null"
}
]
},
"executor": {
"$ref": "#/$defs/ExecutorForPodResponse"
},
"volume": {
"anyOf": [
{
"$ref": "#/$defs/VolumeForPodResponse"
},
{
"type": "null"
}
]
},
"ssh_keys": {
"anyOf": [
{
"items": {
"$ref": "#/$defs/SshKeyForPodResponse"
},
"type": "array"
},
{
"type": "null"
}
],
"title": "Ssh Keys"
},
"notice_period": {
"anyOf": [
{
"$ref": "#/$defs/NoticePeriodResponse"
},
{
"type": "null"
}
]
},
"removal_at": {
"anyOf": [
{
"type": "string",
"format": "date-time"
},
{
"type": "null"
}
],
"title": "Removal At"
},
"skip_add_agent_ssh_key": {
"anyOf": [
{
"type": "boolean"
},
{
"type": "null"
}
],
"title": "Skip Add Agent Ssh Key"
},
"is_manual_rental": {
"anyOf": [
{
"type": "boolean"
},
{
"type": "null"
}
],
"title": "Is Manual Rental"
},
"estimated_ready_seconds": {
"anyOf": [
{
"type": "integer"
},
{
"type": "null"
}
],
"title": "Estimated Ready Seconds",
"description": "Seconds left of the time-to-ready estimate made when the rent was accepted; 0 once it has run over. Only while the pod is PENDING or REBOOT_PENDING."
},
"eta_basis": {
"anyOf": [
{
"type": "string",
"enum": [
"executor_history",
"class_median",
"cold_pull_estimate",
"fleet_default"
]
},
{
"type": "null"
}
],
"title": "Eta Basis",
"description": "What the estimate was made from: this node's recent creates (executor_history), its GPU class (class_median) or the fleet default (fleet_default); cold_pull_estimate when an image pull the node does not have cached was added on top."
},
"phase": {
"anyOf": [
{
"type": "string",
"enum": [
"queued",
"preparing node",
"connecting to node",
"pulling image",
"creating volume",
"starting container",
"mounting encrypted volume",
"configuring ssh"
]
},
{
"type": "null"
}
],
"title": "Phase",
"description": "The step the create has reached, read from the validator's log lines."
},
"rental_tier": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"title": "Rental Tier"
},
"starting_ssh_connect_cmd": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"title": "Starting Ssh Connect Cmd",
"description": "The SSH command the pod will answer on, set while it is PENDING or REBOOT_PENDING once its SSH port is known, before the API reports the pod RUNNING. Null while the port is not known yet, and in every other status (use `ssh_connect_cmd` there). Older servers do not send this field."
},
"create_progress": {
"anyOf": [
{
"$ref": "#/$defs/CreateProgress"
},
{
"type": "null"
}
],
"description": "While the pod is PENDING or REBOOT_PENDING: `stage` (the step the create has reached, same text as `phase`), `message` (the validator's latest create log line, null before the first) and `updated_at` (ISO-8601 UTC, when the backend got that line or entered the stage). Null otherwise."
},
"node_offline": {
"type": "boolean",
"title": "Node Offline",
"description": "The node failed its health checks (not responding) or is marked inactive. The pod keeps its status; it may not be reachable until the node is back.",
"default": false
},
"billing_paused": {
"type": "boolean",
"title": "Billing Paused",
"description": "A billed pod on a node marked inactive or read as down by the validator: each 5-minute charge is $0 until the node is back (see Billing, When a pod is charged).",
"default": false
},
"node_outage": {
"anyOf": [
{
"$ref": "#/$defs/NodeOutage"
},
{
"type": "null"
}
],
"description": "Set while `node_offline` is true: since when, and how the node's outages in the last 30 days ended. History only, not a prediction. Null otherwise."
},
"ssh_unreachable_since": {
"anyOf": [
{
"type": "string",
"format": "date-time"
},
{
"type": "null"
}
],
"title": "Ssh Unreachable Since",
"description": "Set while the pod is RUNNING but its SSH port stopped answering the validator's probe: since when (ISO-8601 UTC). Null otherwise."
},
"pod_status": {
"anyOf": [
{
"$ref": "#/$defs/PodStatusResponse"
},
{
"type": "null"
}
]
},
"provider_discord_connected": {
"anyOf": [
{
"type": "boolean"
},
{
"type": "null"
}
],
"title": "Provider Discord Connected"
},
"last_event": {
"anyOf": [
{
"$ref": "#/$defs/PodEventResponse"
},
{
"type": "null"
}
]
},
"machine_changed_at": {
"anyOf": [
{
"type": "string",
"format": "date-time"
},
{
"type": "null"
}
],
"title": "Machine Changed At"
},
"restore_backup_id": {
"anyOf": [
{
"type": "string",
"format": "uuid"
},
{
"type": "null"
}
],
"title": "Restore Backup Id"
},
"restore_backup_completed_at": {
"anyOf": [
{
"type": "string",
"format": "date-time"
},
{
"type": "null"
}
],
"title": "Restore Backup Completed At"
},
"disk_used_gb": {
"anyOf": [
{
"type": "number"
},
{
"type": "null"
}
],
"title": "Disk Used Gb",
"description": "GB of the node's disk the pod uses: its container, including Docker-in-Docker data, and its /root volume, as last measured. null until measured."
},
"disk_size_gb": {
"anyOf": [
{
"type": "integer"
},
{
"type": "null"
}
],
"title": "Disk Size Gb",
"description": "GB of disk the pod was given: its /root volume plus its pod disk. null on a pod created without a storage limit."
},
"cluster_overlay_ip": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"title": "Cluster Overlay Ip",
"description": "This node's address on the cluster's private overlay — where its peers reach it.",
"readOnly": true
}
},
"type": "object",
"required": [
"executor_id",
"ports_mapping",
"pod_name",
"ssh_connect_cmd",
"gpu_name",
"gpu_count",
"cpu_name",
"cpu_count",
"ram_total",
"status",
"is_favorite",
"removal_scheduled_at",
"termination_hours",
"jupyter_installation_status",
"jupyter_url",
"enable_jupyter",
"enable_volume_encryption",
"volume_encryption_status",
"id",
"executor",
"cluster_overlay_ip"
],
"$defs": {
"ActorMetadata": {
"properties": {
"auth": {
"type": "string",
"title": "Auth"
},
"api_key_id": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"title": "Api Key Id"
},
"api_key_name": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"title": "Api Key Name"
}
},
"type": "object",
"required": [
"auth"
],
"title": "ActorMetadata",
"description": "Who performed the action the event records (DAH-2934).\n\nEvent.user_id is the account; with several API keys on one account it cannot say which of them\nacted. Stamped on insert from the request's auth (see models/event.py); absent on events written\nby workers and the validator consumer."
},
"CreateProgress": {
"properties": {
"stage": {
"type": "string",
"title": "Stage"
},
"message": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"title": "Message"
},
"updated_at": {
"type": "string",
"title": "Updated At"
}
},
"type": "object",
"required": [
"stage",
"message",
"updated_at"
],
"title": "CreateProgress"
},
"ExecutorForPodResponse": {
"properties": {
"id": {
"type": "string",
"format": "uuid",
"title": "Id"
},
"machine_name": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"title": "Machine Name"
},
"price_per_gpu": {
"anyOf": [
{
"type": "number"
},
{
"type": "null"
}
],
"title": "Price Per Gpu"
},
"pending_price_per_hour": {
"anyOf": [
{
"type": "number"
},
{
"type": "null"
}
],
"title": "Pending Price Per Hour"
},
"price_change_effective_date": {
"anyOf": [
{
"type": "string",
"format": "date-time"
},
{
"type": "null"
}
],
"title": "Price Change Effective Date"
},
"executor_ip_address": {
"type": "string",
"title": "Executor Ip Address"
},
"validator_hotkey": {
"type": "string",
"title": "Validator Hotkey"
},
"gpu_count": {
"anyOf": [
{
"type": "integer"
},
{
"type": "null"
}
],
"title": "Gpu Count"
},
"available_gpu_count": {
"anyOf": [
{
"type": "integer"
},
{
"type": "null"
}
],
"title": "Available Gpu Count"
},
"specs": {
"additionalProperties": true,
"type": "object",
"title": "Specs"
},
"active": {
"anyOf": [
{
"type": "boolean"
},
{
"type": "null"
}
],
"title": "Active"
},
"is_responsive": {
"anyOf": [
{
"type": "boolean"
},
{
"type": "null"
}
],
"title": "Is Responsive"
},
"uptime_in_minutes": {
"anyOf": [
{
"type": "integer"
},
{
"type": "null"
}
],
"title": "Uptime In Minutes"
},
"miner_hotkey": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"title": "Miner Hotkey"
},
"location": {
"anyOf": [
{
"$ref": "#/$defs/LocationResponseForExecutor"
},
{
"type": "null"
}
]
},
"is_slow_machine": {
"type": "boolean",
"title": "Is Slow Machine",
"default": false
},
"effective_upload_speed_mbps": {
"anyOf": [
{
"type": "number"
},
{
"type": "null"
}
],
"title": "Effective Upload Speed Mbps"
},
"effective_download_speed_mbps": {
"anyOf": [
{
"type": "number"
},
{
"type": "null"
}
],
"title": "Effective Download Speed Mbps"
},
"interconnect": {
"anyOf": [
{
"$ref": "#/$defs/InterconnectSpec"
},
{
"type": "null"
}
],
"description": "How the GPUs are wired to each other, from the validator's nvidia-smi topo run: nvlink/p2p verdicts, pair counts, PCIe class and the GPU x GPU matrix. null = no scrape has reported it for this host yet. A key the validator wrote in a shape the API cannot type is null on its own; the other keys stay. interconnect.nvlink is the topo -m reading alone; the top-level nvlink is that reading confirmed by the other tables."
},
"tier": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"title": "Tier",
"default": "secure"
},
"offers_bare_metal": {
"type": "boolean",
"title": "Offers Bare Metal",
"description": "The node's provider offers bare-metal rentals (POST /pods/{id}/bare-metal on a whole-node pod).",
"default": false
},
"last_verified_at": {
"anyOf": [
{
"type": "string",
"format": "date-time"
},
{
"type": "null"
}
],
"title": "Last Verified At",
"description": "When the validator last passed this node: the newest positive validation report the backend received about it, UTC. The validator re-checks a listed node about every 15 minutes, so an older time means its recent checks failed or did not arrive. null = no passed check recorded yet."
},
"reliability_score": {
"anyOf": [
{
"type": "number"
},
{
"type": "null"
}
],
"title": "Reliability Score"
},
"node_reliability_pct": {
"anyOf": [
{
"type": "number"
},
{
"type": "null"
}
],
"title": "Node Reliability Pct",
"description": "This node's own reliability score, 0-100. null when no score is available, including while the node is new."
},
"node_reliability_new": {
"type": "boolean",
"title": "Node Reliability New",
"description": "true while the node is new and has no reliability score yet.",
"default": false
},
"provider_spot_avg_uptime_hours": {
"anyOf": [
{
"type": "number"
},
{
"type": "null"
}
],
"title": "Provider Spot Avg Uptime Hours",
"description": "Spot nodes only: average hours this provider's Spot nodes stayed up before going down, over the last 7 days, the same for every Spot node of one provider. null = under 24 h of Spot history."
},
"frequent_rent_failures": {
"type": "boolean",
"title": "Frequent Rent Failures",
"description": "true when, in the last 7 days, at least 2 rents on this node failed to start for the host's reasons and they are at least 30% of its rents; the nodes `reliable_only=true` hides. Judged per node.",
"default": false
},
"provider_rating": {
"anyOf": [
{
"$ref": "#/$defs/ProviderRating"
},
{
"type": "null"
}
]
},
"provider_notes": {
"items": {
"type": "string"
},
"type": "array",
"title": "Provider Notes",
"description": "Notes the provider shows renters on this node, each from a fixed list (for example \"May reboot occasionally\"). Empty = none."
},
"provider_key": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"title": "Provider Key"
},
"gpu_power_limited": {
"anyOf": [
{
"type": "boolean"
},
{
"type": "null"
}
],
"title": "Gpu Power Limited",
"description": "true when a GPU on this node runs under 95 % of its default power limit, so a rental gets less than the card's stock power; false when every GPU that reports a default is at it. null = cannot be judged: no GPU reports both limits, or every GPU that reports sits at Pearl's own temporary cap (70 % of the default) or at the card's driver minimum, which Pearl's cap clamps to on cards whose minimum is above 70 % (RTX 4090, RTX 5090). Such a reading counts as Pearl's while a Lium default-job container is on the node, and also without one when it is under 80 % of the default, since the cap outlives the container; so a host limit at that minimum is not judged in those two cases. A host limit between the minimum and the default is. While Pearl's cap is on and the validator reports the host's own limit under it, that host limit (what the rental gets once Pearl stops) is judged instead."
},
"gpu_power_limit_w": {
"anyOf": [
{
"type": "integer"
},
{
"type": "null"
}
],
"title": "Gpu Power Limit W",
"description": "Power limit in watts of the most reduced GPU (the host's own limit while Pearl's cap is live, when reported); null when gpu_power_limited is null."
},
"gpu_power_limit_default_w": {
"anyOf": [
{
"type": "integer"
},
{
"type": "null"
}
],
"title": "Gpu Power Limit Default W",
"description": "That GPU's default power limit in watts; null when gpu_power_limited is null."
},
"ncu_profiling_enabled": {
"type": "boolean",
"title": "Ncu Profiling Enabled",
"readOnly": true
},
"disk_health_ok": {
"anyOf": [
{
"type": "boolean"
},
{
"type": "null"
}
],
"title": "Disk Health Ok",
"description": "Whether the node's docker root takes writes (DAH-2929). True: no read-only mount over the\ndocker root and the write probe did not fail. False: one of those; the readings behind it are\nin `specs.disk_health`. null: the validator has not probed this node this cycle.",
"readOnly": true
},
"on_infiniband_fabric": {
"type": "boolean",
"title": "On Infiniband Fabric",
"readOnly": true
},
"on_roce_fabric": {
"type": "boolean",
"title": "On Roce Fabric",
"readOnly": true
},
"on_efa_fabric": {
"type": "boolean",
"title": "On Efa Fabric",
"readOnly": true
},
"nvlink": {
"anyOf": [
{
"type": "boolean"
},
{
"type": "null"
}
],
"title": "Nvlink",
"description": "true when the validator saw every GPU pair joined by NVLink and confirmed it: P2P reads OK on every pair, every GPU has its NVLinks up and the topology covers every GPU. false when a pair is not on NVLink or a check contradicts it; null when the topology is unreported or unconfirmed, or the node has one GPU.",
"readOnly": true
}
},
"type": "object",
"required": [
"id",
"executor_ip_address",
"validator_hotkey",
"specs",
"ncu_profiling_enabled",
"disk_health_ok",
"on_infiniband_fabric",
"on_roce_fabric",
"on_efa_fabric",
"nvlink"
],
"title": "ExecutorForPodResponse"
},
"InterconnectSpec": {
"properties": {
"gpu_count": {
"anyOf": [
{
"type": "integer"
},
{
"type": "null"
}
],
"title": "Gpu Count"
},
"gpu_pairs": {
"anyOf": [
{
"type": "integer"
},
{
"type": "null"
}
],
"title": "Gpu Pairs"
},
"nvlink": {
"anyOf": [
{
"type": "boolean"
},
{
"type": "null"
}
],
"title": "Nvlink"
},
"nvlink_links": {
"anyOf": [
{
"type": "integer"
},
{
"type": "null"
}
],
"title": "Nvlink Links"
},
"nvlink_pairs": {
"anyOf": [
{
"type": "integer"
},
{
"type": "null"
}
],
"title": "Nvlink Pairs"
},
"nvlink_active_links": {
"anyOf": [
{
"type": "integer"
},
{
"type": "null"
}
],
"title": "Nvlink Active Links"
},
"pcie_class": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"title": "Pcie Class"
},
"p2p": {
"anyOf": [
{
"type": "boolean"
},
{
"type": "null"
}
],
"title": "P2P"
},
"p2p_pairs": {
"anyOf": [
{
"type": "integer"
},
{
"type": "null"
}
],
"title": "P2P Pairs"
},
"p2p_ok_pairs": {
"anyOf": [
{
"type": "integer"
},
{
"type": "null"
}
],
"title": "P2P Ok Pairs"
},
"matrix": {
"anyOf": [
{
"items": {
"items": {
"type": "string"
},
"type": "array"
},
"type": "array"
},
{
"type": "null"
}
],
"title": "Matrix"
}
},
"type": "object",
"title": "InterconnectSpec",
"description": "How the GPUs of one host are wired to each other, from the validator's `nvidia-smi topo` run.\n\nScraped by lium-io DAH-2922, declared and exposed here (DAH-2923).\n`nvlink` is True only when EVERY GPU pair is joined by NVLink (an HGX board); a box with NVLink\nbridges on some pairs reports False with `nvlink_pairs` < `gpu_pairs`. `p2p` is True only when\nevery pair of `nvidia-smi topo -p2p r` reads OK — False is the host where NCCL cannot start\nwithout NCCL_P2P_DISABLE=1. Both are None on a single-GPU host (nothing to judge) and when the\ntable that answers them could not be read (`interconnect_scrape_error` on MachineSpecs says why).\n`pcie_class` is nvidia-smi's worst class among the non-NVLink pairs (SYS > NODE > PHB > PXB > PIX).\n`matrix` is the GPU x GPU cell table, kept so a client can print the topology."
},
"JupyterInstallationStatus": {
"type": "string",
"enum": [
"INSTALLING",
"SUCCESS",
"FAILED"
],
"title": "JupyterInstallationStatus"
},
"LocationResponseForExecutor": {
"properties": {
"id": {
"type": "string",
"format": "uuid",
"title": "Id"
},
"created_at": {
"type": "string",
"format": "date-time",
"title": "Created At"
},
"updated_at": {
"anyOf": [
{
# --- truncated at 32 KB (46 KB total) ---
# Full source: https://artifacts.apis.io/lium-io/json-schema/lium-io-pod-detail-response-schema.json
Work with this as data
Every JSON Schema here is available over the APIs.io API and to AI agents over MCP.
MCP server
One button, every client — Claude, Cursor, VS Code and the rest.
https://apis.io/mcp
Tools for schemas
4 MCP tools reach this
find_json_schemasBrowse and filter every JSON Schema in the catalog.apis_io_searchSTART HERE — APIs, providers and tags for one query, each with its total.resolveTurn a domain, URL or GitHub org into the provider it belongs to.find_cohortsEvery scored population of providers in the catalog.
Call it yourself
curl for this page
This JSON Schema
curl "https://apis.io/api/v1/json-schemas/lium-io-pod-detail-response"
All schemas
curl "https://apis.io/api/v1/json-schemas?limit=25"
Calls without X-API-Key need no key. Calls with it need a paid plan — get a key.
Get an API key
Free tier, no form to fill in. Signing in shares your email address with us — we store it to create your key and to recognise you if you sign in with another provider. By signing up you are subscribing to the APIs.io and API Evangelist newsletters — you can unsubscribe from any issue. See our Privacy Policy and Terms.
A second provider on the same verified email joins the account you already have.