Prime Intellect

Prime Intellect is a San Francisco–based startup building an open and decentralized stack for developing and improving AI agents through reinforcement learning. The company combines a 50+ provider GPU compute marketplace (single-node and 1–256 GPU multi-node clusters across H100, H200, B200, and B300), a managed RL post-training service ("Lab") that fine-tunes Qwen3.5, Llama, NVIDIA Nemotron, and OpenAI gpt-oss base models against a 2,500+ environment library (the Environments Hub), secure remote Python sandboxes for agent code execution, OpenAI-compatible inference at api.pinference.ai with LoRA adapter deployments, and an evaluations service that runs against the same environments. Prime Intellect is also the publisher of INTELLECT-3 (100B+ MoE trained with RL), INTELLECT-2 (32B trained via distributed RL), and the SYNTHETIC-2 dataset of four million reasoning traces, and develops the open-source `verifiers` RL library and the `prime-rl` async RL training framework. Everything is exposed via a single bearer-token REST API at api.primeintellect.ai, a `prime` Python CLI/SDK (PyPI: `prime`, lightweight `prime-sandboxes`), and the OpenAI protocol on the inference plane.

6 APIs 0 Features

AIArtificial IntelligenceReinforcement LearningGPU ComputeDecentralized ComputeFoundation ModelsInferenceSandboxesTrainingEnvironmentsEvaluationsLoRAOpen SourceAgents

APIs

Prime Intellect Compute API

Programmatic access to the Prime Intellect compute marketplace. Query GPU availability across 50+ providers, provision and manage on-demand or multi-node GPU pods (1-256 GPUs ac...

Prime Intellect Sandbox API

Secure remote code-execution sandboxes for AI agents. Create and manage Python-3.11-slim-based sandboxes, expose ports for HTTP services running inside the sandbox, open SSH ses...

Prime Intellect Training API

Hosted reinforcement-learning post-training service ("Lab"). Launch dedicated training runs against a catalog of base models — Qwen3.5 (0.8B through 397B-A17B), Llama 1B/3B Inst...

Prime Intellect Evaluations API

Create, manage, and submit samples to evaluations against the Prime Intellect Environments Hub (2,500+ RL environments). Supports both client-driven evaluations (push samples, f...

Prime Intellect Inference API

OpenAI-compatible inference API for hosted frontier and open models served at api.pinference.ai. Supports streaming chat completions, the full set of OpenAI parameters (temperat...

Prime Intellect Platform API

Account, team, secrets, image, and cluster administration. Whoami and team membership, secret CRUD for use inside pods and sandboxes, Docker image builds with public/private vis...

Semantic Vocabularies

Prime Intellect Context

0 classes · 8 properties

JSON-LD

Resources

Documentation

APIReference

Authentication

HuggingFace

Sources

aid: prime-intellect
url: https://raw.githubusercontent.com/api-evangelist/prime-intellect/refs/heads/main/apis.yml
apis:
  - aid: prime-intellect:prime-intellect-compute-api
    name: Prime Intellect Compute API
    tags:
      - GPU Compute
      - Marketplace
      - Pods
      - Disks
      - SSH Keys
      - Availability
    humanURL: https://docs.primeintellect.ai/api-reference/managing-pods
    baseURL: https://api.primeintellect.ai
    properties:
      - url: https://docs.primeintellect.ai/api-reference/introduction
        type: Documentation
      - url: https://docs.primeintellect.ai/api-reference/managing-pods
        type: Documentation
      - url: https://docs.primeintellect.ai/api-reference/managing-disks
        type: Documentation
      - url: https://docs.primeintellect.ai/api-reference/check-gpu-availability
        type: Documentation
      - url: openapi/prime-intellect-compute-api-openapi.yml
        type: OpenAPI
      - url: json-schema/prime-intellect-pod-schema.json
        type: JSONSchema
      - url: json-ld/prime-intellect-context.jsonld
        type: JSONLD
    description: >-
      Programmatic access to the Prime Intellect compute marketplace. Query GPU availability across 50+ providers,
      provision and manage on-demand or multi-node GPU pods (1-256 GPUs across H100, H200, B200, B300, A100), attach
      persistent network disks, and manage SSH keys for pod access. Backs the `prime pods`, `prime availability`, and
      `prime disks` CLI commands.
  - aid: prime-intellect:prime-intellect-sandbox-api
    name: Prime Intellect Sandbox API
    tags:
      - Sandbox
      - Code Execution
      - Agents
      - Tunnels
    humanURL: https://docs.primeintellect.ai/sandboxes/overview
    baseURL: https://api.primeintellect.ai
    properties:
      - url: https://docs.primeintellect.ai/sandboxes/overview
        type: Documentation
      - url: https://docs.primeintellect.ai/sandboxes/sdk
        type: Documentation
      - url: https://docs.primeintellect.ai/sandboxes/cli
        type: Documentation
      - url: https://docs.primeintellect.ai/sandboxes/tunnel
        type: Documentation
      - url: openapi/prime-intellect-sandbox-api-openapi.yml
        type: OpenAPI
      - url: json-schema/prime-intellect-sandbox-schema.json
        type: JSONSchema
    description: >-
      Secure remote code-execution sandboxes for AI agents. Create and manage Python-3.11-slim-based sandboxes, expose
      ports for HTTP services running inside the sandbox, open SSH sessions, pull rich error context for failures, and
      provision reverse tunnels back into a sandbox from an external network. Available as a lightweight
      `prime-sandboxes` Python package (~50KB) in addition to the full `prime` CLI/SDK.
  - aid: prime-intellect:prime-intellect-training-api
    name: Prime Intellect Training API
    tags:
      - Reinforcement Learning
      - Training
      - Fine-Tuning
      - Foundation Models
      - Billing
    humanURL: https://docs.primeintellect.ai/hosted-training/what-is-lab
    baseURL: https://api.primeintellect.ai
    properties:
      - url: https://docs.primeintellect.ai/hosted-training/what-is-lab
        type: Documentation
      - url: https://docs.primeintellect.ai/hosted-training/getting-started
        type: Documentation
      - url: https://docs.primeintellect.ai/hosted-training/end-to-end-run
        type: Documentation
      - url: https://docs.primeintellect.ai/hosted-training/models-and-pricing
        type: Pricing
      - url: openapi/prime-intellect-training-api-openapi.yml
        type: OpenAPI
      - url: json-schema/prime-intellect-training-run-schema.json
        type: JSONSchema
    description: >-
      Hosted reinforcement-learning post-training service ("Lab"). Launch dedicated training runs against a catalog of
      base models — Qwen3.5 (0.8B through 397B-A17B), Llama 1B/3B Instruct, NVIDIA Nemotron 30B/120B, OpenAI gpt-oss
      20B/120B — drive them with environments from the Environments Hub, and retrieve token-level usage + USD cost per
      run from the billing endpoints. Pricing is per-million-token across separate input, output, and training meters.
  - aid: prime-intellect:prime-intellect-evaluations-api
    name: Prime Intellect Evaluations API
    tags:
      - Evaluations
      - Benchmarks
      - Reinforcement Learning
      - Environments
    humanURL: https://docs.primeintellect.ai/tutorials-environments/hosted-evaluations
    baseURL: https://api.primeintellect.ai
    properties:
      - url: https://docs.primeintellect.ai/tutorials-environments/environments
        type: Documentation
      - url: https://docs.primeintellect.ai/tutorials-environments/evaluating
        type: Documentation
      - url: https://docs.primeintellect.ai/tutorials-environments/hosted-evaluations
        type: Documentation
      - url: openapi/prime-intellect-evaluations-api-openapi.yml
        type: OpenAPI
      - url: json-schema/prime-intellect-evaluation-schema.json
        type: JSONSchema
    description: >-
      Create, manage, and submit samples to evaluations against the Prime Intellect Environments Hub (2,500+ RL
      environments). Supports both client-driven evaluations (push samples, finalize, retrieve) and fully hosted
      evaluations executed by Prime Intellect with logs, cancellation, and selectable inference models. Backs `prime
      eval` CLI commands.
  - aid: prime-intellect:prime-intellect-inference-api
    name: Prime Intellect Inference API
    tags:
      - Inference
      - OpenAI Compatible
      - Foundation Models
      - LLM
    humanURL: https://docs.primeintellect.ai/inference/overview
    baseURL: https://api.pinference.ai/api/v1
    properties:
      - url: https://docs.primeintellect.ai/inference/overview
        type: Documentation
      - url: https://docs.primeintellect.ai/inference/usage
        type: Documentation
      - url: https://docs.primeintellect.ai/inference/adapter-deployments
        type: Documentation
      - url: openapi/prime-intellect-inference-api-openapi.yml
        type: OpenAPI
    description: >-
      OpenAI-compatible inference API for hosted frontier and open models served at api.pinference.ai. Supports
      streaming chat completions, the full set of OpenAI parameters (temperature, top_p, max_tokens, logprobs), and
      returns a `usage` object with input/output token counts and USD cost on every response. LoRA adapters can be
      served alongside base models via 1-click deployments.
  - aid: prime-intellect:prime-intellect-platform-api
    name: Prime Intellect Platform API
    tags:
      - Account
      - Teams
      - Secrets
      - Images
      - Administration
    humanURL: https://docs.primeintellect.ai/api-reference/teams
    baseURL: https://api.primeintellect.ai
    properties:
      - url: https://docs.primeintellect.ai/api-reference/teams
        type: Documentation
      - url: https://docs.primeintellect.ai/api-reference/api-keys
        type: Authentication
      - url: openapi/prime-intellect-platform-api-openapi.yml
        type: OpenAPI
    description: >-
      Account, team, secrets, image, and cluster administration. Whoami and team membership, secret CRUD for use inside
      pods and sandboxes, Docker image builds with public/private visibility, registry credential checks, FRP plugin
      validation, and admin-only cluster + node-log endpoints.
name: Prime Intellect
description: >-
  Prime Intellect is a San Francisco–based startup building an open and decentralized stack for developing and improving
  AI agents through reinforcement learning. The company combines a 50+ provider GPU compute marketplace (single-node and
  1–256 GPU multi-node clusters across H100, H200, B200, and B300), a managed RL post-training service ("Lab") that
  fine-tunes Qwen3.5, Llama, NVIDIA Nemotron, and OpenAI gpt-oss base models against a 2,500+ environment library (the
  Environments Hub), secure remote Python sandboxes for agent code execution, OpenAI-compatible inference at
  api.pinference.ai with LoRA adapter deployments, and an evaluations service that runs against the same environments.
  Prime Intellect is also the publisher of INTELLECT-3 (100B+ MoE trained with RL), INTELLECT-2 (32B trained via
  distributed RL), and the SYNTHETIC-2 dataset of four million reasoning traces, and develops the open-source
  `verifiers` RL library and the `prime-rl` async RL training framework. Everything is exposed via a single bearer-token
  REST API at api.primeintellect.ai, a `prime` Python CLI/SDK (PyPI: `prime`, lightweight `prime-sandboxes`), and the
  OpenAI protocol on the inference plane.
tags:
  - AI
  - Artificial Intelligence
  - Reinforcement Learning
  - GPU Compute
  - Decentralized Compute
  - Foundation Models
  - Inference
  - Sandboxes
  - Training
  - Environments
  - Evaluations
  - LoRA
  - Open Source
  - Agents
kind: contract
image: https://kinlane-images.s3.amazonaws.com/shared/apis-json/apis-json-logo.jpg
access: 3rd-Party
type: Index
position: Provider
specificationVersion: '0.20'
created: '2026-05-25'
modified: '2026-05-25'
common:
  - type: Website
    url: https://www.primeintellect.ai
  - type: Documentation
    url: https://docs.primeintellect.ai
  - type: Portal
    url: https://app.primeintellect.ai
  - type: APIReference
    url: https://docs.primeintellect.ai/api-reference/introduction
  - type: OpenAPI
    url: https://api.primeintellect.ai/openapi.json
  - type: Authentication
    url: https://docs.primeintellect.ai/api-reference/api-keys
  - type: CLI
    url: https://docs.primeintellect.ai/cli-reference/introduction
  - type: SDK
    url: https://github.com/PrimeIntellect-ai/prime
    name: Prime CLI & Python SDK
  - type: SDK
    url: https://pypi.org/project/prime/
    name: prime (PyPI)
  - type: SDK
    url: https://pypi.org/project/prime-sandboxes/
    name: prime-sandboxes (PyPI)
  - type: OpenSource
    url: https://github.com/PrimeIntellect-ai/verifiers
    name: verifiers — RL environments + evals library
  - type: OpenSource
    url: https://github.com/PrimeIntellect-ai/prime-rl
    name: prime-rl — async RL training framework
  - type: OpenSource
    url: https://github.com/PrimeIntellect-ai/renderers
    name: renderers — programmable chat templates
  - type: OpenSource
    url: https://github.com/PrimeIntellect-ai/research-environments
    name: research-environments
  - type: OpenSource
    url: https://github.com/PrimeIntellect-ai/community-environments
    name: community-environments
  - type: GitHub
    url: https://github.com/PrimeIntellect-ai
  - type: HuggingFace
    url: https://huggingface.co/PrimeIntellect
  - type: Blog
    url: https://www.primeintellect.ai/blog
  - type: Pricing
    url: https://docs.primeintellect.ai/hosted-training/models-and-pricing
  - type: Plans
    url: plans/prime-intellect-plans-pricing.yml
  - type: RateLimits
    url: rate-limits/prime-intellect-rate-limits.yml
  - type: FinOps
    url: finops/prime-intellect-finops.yml
  - type: Discord
    url: https://discord.gg/primeintellect
  - type: Twitter
    url: https://x.com/PrimeIntellect
  - type: LinkedIn
    url: https://www.linkedin.com/company/primeintellect-ai
  - type: Events
    url: https://luma.com/primeintellect
  - type: Careers
    url: https://jobs.ashbyhq.com/PrimeIntellect
  - type: Contact
    url: https://www.primeintellect.ai/contact
maintainers:
  - FN: Kin Lane
    email: kin@apievangelist.com