Workloads

CreateWorkload

POST//v1/workloads

Body

adapter_artifactsstring[]

allowed_clustersstring[]

allowed_regionsstring[]

cluster

cluster_floorsobject

deadline_at

desired_scaleintegerdefault:1

enabledbooleandefault:true

gang_sizeintegerdefault:1

kindstringdefault:"service"

labelsobject

max_clusters

max_scaleintegerdefault:1

min_scaleintegerdefault:1

namestringrequired

note

overridesobject

preemptiblebooleandefault:false

priority_class

projectstringrequired

servingobjectrequired

strategystringdefault:"pack"

Response

adapter_artifact_idsstring[]required

artifact_tag

cluster_floorsobjectrequired

created_atstring<date-time>required

deadline_at

deploymentsobject[]required

finished_at

gang_sizeintegerrequired

idstringrequired

kindstringrequired

labelsobjectrequired

max_clusters

namestringrequired

overridesobjectrequired

pausedbooleanrequired

phasestringrequired

pinned_cluster_id

preemptiblebooleanrequired

priority_classstringrequired

project_idstringrequired

revisionstringrequired

scaleobjectrequired

served_model_namestringrequired

servingobjectrequired

serving_digeststringrequired

started_at

strategystringrequired

updated_at

curl -X POST '//v1/workloads' \ -H 'Accept: application/json' \ -H 'Content-Type: application/json' \ -d '{ "adapter_artifacts": [ "string" ], "allowed_clusters": [ "string" ], "allowed_regions": [ "string" ], "cluster": null, "cluster_floors": {}, "deadline_at": null, "desired_scale": 1, "enabled": true, "gang_size": 1, "kind": "service", "labels": {}, "max_clusters": null, "max_scale": 1, "min_scale": 1, "name": "string", "note": null, "overrides": {}, "preemptible": false, "priority_class": null, "project": "string", "serving": { "artifact": "string", "cache": { "kv_cache_dtype": null, "kv_cache_fraction": null, "prefix_cache_config": {}, "prefix_cache_enabled": true }, "capabilities": { "asserted_at": null, "asserted_by": null, "input_modalities": [ "string" ], "parallel_tool_calls": false, "prompt_cache_mode": "off", "prompt_cache_ttl_seconds": null, "reasoning_supported": false, "source_uri": null, "streaming_supported": true, "structured_output_modes": [ "string" ], "tool_calling_supported": false, "tool_choice_modes": [ "string" ] }, "context_length": 0, "dtype": null, "engine": "string", "engine_args": {}, "engine_runtime_image": null, "engine_runtime_version": "string", "frontend_image": null, "imported_args": {}, "max_batch_tokens": null, "overlay": { "dynamo": {}, "pod_spec": {} }, "parallelism": { "expert": 1, "pipeline": 1, "tensor": 1 }, "quantization": null, "resources": { "cpu_cores": null, "disk_gib": null, "memory_mib": null }, "router": { "kv_overlap_score_credit": null, "mode": null, "queue_policy": null, "queue_threshold": null }, "serving": { "context_overflow": "reject", "diffusion": null, "max_images_per_request": null, "max_videos_per_request": null, "output_modality": null, "reasoning": { "default_on": false, "parser": null, "suppress_eos_until_think_end": false, "think_begin_token": null, "think_end_token": null, "transport": "reasoning_content" }, "stream_usage": "final_chunk", "structured_output_backend": null, "tool_call_parser": null }, "shape": { "device_class": "string", "device_count": 1, "device_memory_mib_min": 0, "requires_single_topology_domain": false, "topology_domain_kind": null }, "topology": { "roles": [ { "engine_args": {}, "max_batch_tokens": null, "role": "string", "weight": 1 } ], "serving_topology": "aggregated" } }, "strategy": "pack" }'

{ "adapter_artifact_ids": [ "string" ], "artifact_tag": null, "cluster_floors": {}, "created_at": "2023-01-01T00:00:00Z", "deadline_at": null, "deployments": [ { "admission_phase": "string", "admission_reason": null, "admitted_at": null, "authority": "string", "cluster_id": "string", "cluster_name": "string", "desired_replicas": 0, "enabled": true, "generation": 0, "id": "string", "live_digests": {}, "max_replicas": 0, "min_replicas": 0, "name": "string", "placed_at": null, "queued_at": null, "region": null, "replicas": { "at_current_generation": 0, "failed": 0, "pending": 0, "ready": 0, "running": 0, "total": 0 }, "status": null } ], "finished_at": null, "gang_size": 0, "id": "string", "kind": "string", "labels": {}, "max_clusters": null, "name": "string", "overrides": {}, "paused": true, "phase": "string", "pinned_cluster_id": null, "preemptible": true, "priority_class": "string", "project_id": "string", "revision": "string", "scale": { "desired": 0, "max": 0, "min": 0, "placed": 0, "ready": 0, "unit": "string" }, "served_model_name": "string", "serving": { "artifact_id": "string", "cache": { "kv_cache_dtype": null, "kv_cache_fraction": null, "prefix_cache_config": {}, "prefix_cache_enabled": true }, "capabilities": { "asserted_at": null, "asserted_by": null, "input_modalities": [ "string" ], "parallel_tool_calls": false, "prompt_cache_mode": "off", "prompt_cache_ttl_seconds": null, "reasoning_supported": false, "source_uri": null, "streaming_supported": true, "structured_output_modes": [ "string" ], "tool_calling_supported": false, "tool_choice_modes": [ "string" ] }, "context_length": 0, "dtype": null, "engine": "string", "engine_args": {}, "engine_runtime_image": null, "engine_runtime_version": "string", "frontend_image": null, "imported_args": {}, "max_batch_tokens": null, "overlay": { "dynamo": {}, "pod_spec": {} }, "parallelism": { "expert": 1, "pipeline": 1, "tensor": 1 }, "quantization": null, "resolved_engine_args": {}, "resources": { "cpu_cores": null, "disk_gib": null, "memory_mib": null }, "router": { "kv_overlap_score_credit": null, "mode": null, "queue_policy": null, "queue_threshold": null }, "serving": { "context_overflow": "reject", "diffusion": null, "max_images_per_request": null, "max_videos_per_request": null, "output_modality": null, "reasoning": { "default_on": false, "parser": null, "suppress_eos_until_think_end": false, "think_begin_token": null, "think_end_token": null, "transport": "reasoning_content" }, "stream_usage": "final_chunk", "structured_output_backend": null, "tool_call_parser": null }, "shape": { "device_class": "string", "device_count": 1, "device_memory_mib_min": 0, "requires_single_topology_domain": false, "topology_domain_kind": null }, "topology": { "group_devices": 0, "roles": [ { "engine_args": {}, "max_batch_tokens": null, "role": "string", "weight": 1 } ], "serving_topology": "aggregated" } }, "serving_digest": "string", "started_at": null, "strategy": "string", "updated_at": null }