> ## Documentation Index > Fetch the complete documentation index at: https://docs.veri.studio/llms.txt > Use this file to discover all available pages before exploring further. # Create deployment ## OpenAPI ````yaml /api-reference/openapi.json post /v1/deployments openapi: 3.1.0 info: title: Veri API description: >- REST API for the Veri RL post-training platform. All requests require a Bearer API key (`vk_` prefix). license: name: '' version: 0.1.0 servers: - url: https://api.veri.studio description: Production security: [] tags: - name: Training jobs description: Create, monitor, and manage training jobs. - name: Datasets description: Upload and connect training datasets. - name: Reward functions description: Upload and manage reward functions. - name: Deployments description: Serve trained models and run inference. - name: Evaluations description: Score models against datasets and benchmarks. - name: Volumes description: Persistent file storage mounted into jobs. - name: Models description: Custom model registry deployments serve from. - name: Regions description: Discover available launch regions. - name: GPU description: Live GPU availability by provider and region. - name: Code artifacts description: Upload custom training script bundles. - name: Billing description: Credit balance and transaction history. - name: API keys description: Create and revoke API keys. - name: Account description: The authenticated caller's identity. - name: SSH keys description: Manage SSH key pairs for compute access. - name: Settings description: Account-level integrations (Weights & Biases). paths: /v1/deployments: post: tags: - Deployments summary: Create deployment operationId: create requestBody: content: application/json: schema: $ref: '#/components/schemas/DeploymentCreate' required: true responses: '200': description: The created deployment content: application/json: schema: $ref: '#/components/schemas/DeploymentResponse' '422': description: Request body failed validation security: - bearerAuth: [] components: schemas: DeploymentCreate: type: object required: - model - name - gpu properties: model: type: string source: type: string name: type: string gpu: $ref: '#/components/schemas/GpuConfig' provider: type: - string - 'null' engine: type: - string - 'null' description: >- Inference engine: "vllm" (default) or "max" (Modular MAX). Omitted => vllm. num_replicas: type: - integer - 'null' format: int32 description: >- VS-341: number of GPU-box replicas to launch. Omitted => 1, which is byte-for-byte today's single-box behavior. Validated 1..=8 at create. Sugar for min_replicas == max_replicas == N (the CLI --replicas flag). min_replicas: type: - integer - 'null' format: int32 description: >- Scaling foundation (ยง10): replica-count bounds. min == max => fixed N (autoscaling off โ€” there is no boolean flag). Omitted => both default to num_replicas (fixed N). min = 0 arms scale-to-zero (VS-313): the fleet parks after the idle window and wakes on the next request. max_replicas: type: - integer - 'null' format: int32 concurrency_target: type: - number - 'null' format: double description: >- Per-replica concurrency setpoint the (VS-342) autoscaler targets: desired = ceil(total_ongoing / concurrency_target), clamped to [min_replicas, max_replicas]. Stored as target_ongoing_requests. Omitted => 8 (throughput-biased; lower it for latency-sensitive chat). scale_to_zero_window_seconds: type: - integer - 'null' format: int32 description: >- VS-313: idle seconds before a min_replicas=0 deployment parks (scaled_to_zero). Omitted => 3600. Floor 300 (Fireworks' convention). idle_delete_after_days: type: - integer - 'null' format: int32 description: >- VS-313: optional GC โ€” a deployment parked at zero with no traffic for this many days goes terminal (stopped). Omitted => kept forever. DeploymentResponse: type: object required: - object - id - status - name - model - source - created_at - updated_at properties: object: type: string id: type: string status: $ref: '#/components/schemas/DeploymentStatus' name: type: string model: type: string source: type: string source_model_id: type: - string - 'null' description: |- For a custom_model deployment, the saved model this was created from (provenance). Null for other sources. endpoint_url: type: - string - 'null' gpu: oneOf: - type: 'null' - $ref: '#/components/schemas/GpuInfo' provider: type: - string - 'null' engine: type: - string - 'null' num_replicas: type: integer format: int32 description: >- VS-341: DESIRED replica count (GPU boxes) behind this deployment. The observed count is ready_replicas โ€” the K8s spec/status split; the CLI renders "ready/desired" (e.g. 2/3). ready_replicas: type: integer format: int32 description: 'Observed: replicas currently serving with a routable endpoint.' min_replicas: type: integer format: int32 description: Replica-count bounds. min == max => fixed N (autoscaling off). max_replicas: type: integer format: int32 concurrency_target: type: - number - 'null' format: double description: >- Per-replica concurrency setpoint (stored as target_ongoing_requests). scale_to_zero_window_seconds: type: - integer - 'null' format: int32 description: 'VS-313: idle window before a min=0 deployment parks (None = 3600s).' idle_delete_after_days: type: - integer - 'null' format: int32 description: 'VS-313: parked-deployment GC in days (None = kept forever).' cost_per_hour_usd: type: - number - 'null' format: double total_cost_usd: type: - number - 'null' format: double total_requests: type: integer format: int32 error: type: - string - 'null' started_at: type: - string - 'null' format: date-time stopped_at: type: - string - 'null' format: date-time created_at: type: string format: date-time updated_at: type: string format: date-time GpuConfig: type: object required: - gpu_type - gpu_count properties: gpu_type: type: string gpu_count: type: integer format: int32 DeploymentStatus: type: string enum: - queued - provisioning - serving - unhealthy - stopped - failed - scaled_to_zero - waking GpuInfo: type: object description: |- Pydantic `GPUInfo` (with `type` field). Rust uses `gpu_type` internally but serializes as "type" for JSON parity. required: - type - count properties: type: type: string count: type: integer format: int32 securitySchemes: bearerAuth: type: http scheme: bearer bearerFormat: API key description: API key with the `vk_` prefix. Create one from the dashboard. ````