Prime Intellect · Capability

Prime Intellect Inference API — Chat

Prime Intellect Inference API — Chat. 2 operations. Self-contained Naftiko capability covering one Prime Intellect business surface.

Prime Intellect Inference API — Chat is a Naftiko capability published by Prime Intellect, one of 13 capabilities the APIs.io network indexes for this provider. It bundles 2 operations across the POST and GET methods.

The capability includes 1 read-only operation and 1 state-changing operation. Lead operation: OpenAI-compatible chat completion with streaming and usage metadata. Can be deployed as a REST endpoint, MCP tool, or Agent Skill via Naftiko.

Tagged areas include Prime Intellect and Inference.

Run with Naftiko Prime IntellectInference

What You Can Do

POST
Createchatcompletion — OpenAI-compatible chat completion with streaming and usage metadata.
/v1/chat/completions
GET
Listmodels — List inference models available on api.pinference.ai.
/v1/models

MCP Tools

prime-intellect-createchatcompletion

OpenAI-compatible chat completion with streaming and usage metadata.

prime-intellect-listmodels

List inference models available on api.pinference.ai.

read-only idempotent

Capability Spec

inference-chat.yaml Raw ↑ 
naftiko: 1.0.0-alpha2
info:
  label: Prime Intellect Inference API — Chat
  description: Prime Intellect Inference API — Chat. 2 operations. Self-contained Naftiko capability covering one Prime Intellect
    business surface.
  tags:
  - Prime Intellect
  - Inference
  created: '2026-05-25'
  modified: '2026-05-25'
binds:
- namespace: env
  keys:
    PRIME_API_KEY: PRIME_API_KEY
capability:
  consumes:
  - type: http
    namespace: inference-chat
    baseUri: https://api.pinference.ai/api/v1
    description: Prime Intellect Inference API — Chat business capability. Self-contained, no shared references.
    resources:
    - name: chat-completions
      path: /chat/completions
      operations:
      - name: createchatcompletion
        method: POST
        description: OpenAI-compatible chat completion with streaming and usage metadata.
        outputRawFormat: json
        outputParameters:
        - name: result
          type: object
          value: $.
        inputParameters:
        - name: body
          in: body
          type: object
          description: Request body (JSON).
          required: true
    - name: models
      path: /models
      operations:
      - name: listmodels
        method: GET
        description: List inference models available on api.pinference.ai.
        outputRawFormat: json
        outputParameters:
        - name: result
          type: object
          value: $.
    authentication:
      type: bearer
      value: '{{env.PRIME_API_KEY}}'
      placement: header
  exposes:
  - type: rest
    namespace: inference-chat-rest
    port: 8080
    description: REST adapter for Prime Intellect Inference API — Chat.
    resources:
    - path: /v1/chat/completions
      name: chat-completions
      description: REST surface for chat-completions.
      operations:
      - method: POST
        name: createchatcompletion
        description: OpenAI-compatible chat completion with streaming and usage metadata.
        call: inference-chat.createchatcompletion
        with:
          body: rest.body
        outputParameters:
        - type: object
          mapping: $.
    - path: /v1/models
      name: models
      description: REST surface for models.
      operations:
      - method: GET
        name: listmodels
        description: List inference models available on api.pinference.ai.
        call: inference-chat.listmodels
        with: {}
        outputParameters:
        - type: object
          mapping: $.
  - type: mcp
    namespace: inference-chat-mcp
    port: 9090
    transport: http
    description: MCP adapter for Prime Intellect Inference API — Chat. One tool per consumed operation, routed inline through
      this capability's consumes block.
    tools:
    - name: prime-intellect-createchatcompletion
      description: OpenAI-compatible chat completion with streaming and usage metadata.
      hints:
        readOnly: false
        destructive: false
        idempotent: false
      call: inference-chat.createchatcompletion
      with:
        body: tools.body
      outputParameters:
      - type: object
        mapping: $.
    - name: prime-intellect-listmodels
      description: List inference models available on api.pinference.ai.
      hints:
        readOnly: true
        destructive: false
        idempotent: true
      call: inference-chat.listmodels
      with: {}
      outputParameters:
      - type: object
        mapping: $.