slug: prime-intellect
provider: Prime Intellect
generated_by: planning/capability-mapping/scripts/classify_capabilities.py
model: claude-opus-5
frame:
- Education
min_confidence: 0.7
capability_model:
  source: https://github.com/vincentmakes/turbo-ea-capabilities
  license: CC-BY-4.0
  attribution: Turbo EA Capabilities by Vincent Verdet — Turbo EA, https://github.com/vincentmakes/turbo-ea-capabilities,
    CC BY 4.0
  notice: NOTICE
edge_count: 6
edges:
- tag: Pods
  spec_file: prime-intellect-pods-api-openapi.yml
  capability_id: BC-600.50
  capability_id_l1: BC-600
  capability_name: IT Infrastructure Management
  confidence: 0.8
  evidence: POST /api/v1/pods/  Create Pod; 'on-demand and multi-node GPU pod lifecycle, persistent network-attached
    disks'
  reason: Provisioning, monitoring and deletion of GPU compute pods is compute/cloud infrastructure management.
- tag: Inference
  spec_file: prime-intellect-inference-api-openapi.yml
  capability_id: BC-610.60
  capability_id_l1: BC-610
  capability_name: Artificial Intelligence Management
  confidence: 0.75
  evidence: GET /models  Prime Intellect List Inference Models; POST /chat/completions  Create Chat Completion
  reason: Model listing and chat-completion inference is AI/ML model serving — the operational side of
    AI model lifecycle management. No business-domain reading fits.
- tag: admin-clusters
  spec_file: prime-intellect-admin-clusters-api-openapi.yml
  capability_id: BC-600.50
  capability_id_l1: BC-600
  capability_name: IT Infrastructure Management
  confidence: 0.75
  evidence: GET /api/admin/clusters  List Clusters; GET /api/admin/clusters/{cluster_id}/node-logs  List
    Cluster Node Logs
  reason: Administrative inventory, node inspection and certificate issuance for GPU clusters is compute
    infrastructure management.
- tag: training
  spec_file: prime-intellect-training-api-openapi.yml
  capability_id: BC-610.60
  capability_id_l1: BC-610
  capability_name: Artificial Intelligence Management
  confidence: 0.75
  evidence: POST /api/v1/training/runs  create_dedicated_run  'Create Dedicated Run'
  reason: Creating and deleting dedicated model training runs on GPU clusters is ML model lifecycle /
    MLOps execution, which maps to Artificial Intelligence Management.
- tag: Secrets
  spec_file: prime-intellect-secrets-api-openapi.yml
  capability_id: BC-4210.60
  capability_id_l1: BC-4210
  capability_name: Configuration & Secrets Management
  confidence: 0.7
  evidence: POST /api/v1/secrets/  Create Secret; SecretCreateRequest, SecretUpdateRequest
  reason: CRUD over stored secrets consumed by workloads matches management of runtime configuration values
    and secrets.
- tag: evals
  spec_file: prime-intellect-evals-api-openapi.yml
  capability_id: BC-610.60
  capability_id_l1: BC-610
  capability_name: Artificial Intelligence Management
  confidence: 0.7
  evidence: POST /api/v1/evaluations/  Create Evaluation; POST /api/v1/evaluations/{evaluation_id}/samples  Push
    Samples; HostedEvalConfig
  reason: Creating evaluations, pushing sample results and finalizing them is model evaluation within
    the AI/ML model lifecycle (MLOps). Despite the word 'evaluation', nothing indicates academic assessment
    — the vendor is an AI compute/RL platform.