Surge AI

Surge AI is a human-data company that provides large-scale, expert-quality labeled data for training and evaluating frontier AI models. The product surface spans RL environments and agents (rich, complex environments that challenge agentic models), rubrics and verifiers (scoring systems for AI outputs), RLHF (preference and reward data), SFT (foundational skill demonstrations), human evaluation, expert professional domains, internationalization across 70+ languages, multimodal (image, audio, video) data, and off-the-shelf datasets. Surge ships an official Python SDK (surge-python) wrapping the Surge API, with API-key authentication, and exposes the dashboard and API reference at app.surgehq.ai. Public datasets published by Surge include the toxicity dataset (the world's largest social-media toxicity dataset).

11 APIs 11 Features

Human DataRLHFSFTRubricsVerifiersRL EnvironmentsMultimodalInternationalizationLabeling

Surge AI publishes 11 APIs on the APIs.io network. Tagged areas include Human Data, RLHF, SFT, Rubrics, and Verifiers.

Surge AI’s developer surface includes developer portal, documentation, API reference, authentication, signup flow, developer console, SDKs, and 6 more developer resources.

Python SDK

Official surge-python SDK on PyPI, MIT-licensed, Python 3.10+.

RL Environments and Agents

Complex environments that challenge agentic models.

Rubrics and Verifiers

Scoring systems for AI outputs across domains.

RLHF and SFT

Preference, reward, and demonstration data for foundation-model training.

Human Evaluation

Expert workforce grades AI output quality.

Expert Professional Domains

Specialized expertise across finance, law, medicine, and more.

70+ Languages

Internationalization coverage spanning more than 70 languages.

Multimodal Data

Image, audio, and video collection and labeling.

Off-The-Shelf Datasets

Pre-built datasets available for licensing.

Open Datasets

Public releases including the world's largest social-media toxicity dataset.

Use Cases

Frontier Model RLHF

Preference and reward data for reinforcement learning from human feedback.

Supervised Fine-Tuning

Demonstration data for SFT across professional domains.

Agentic Evals

Benchmark agents in complex RL environments with structured rubrics.

Multilingual Model Evaluation

Evaluate model quality across 70+ languages.

Trust and Safety Research

Use the toxicity dataset and human evaluation pipelines for trust and safety work.

Integrations

Python SDK

Programmatic integration via the official surge-python SDK.

API Key Authentication

Standard API-key auth (SURGE_API_KEY env var or programmatic configuration).

Custom Project Blueprints

Use Surge blueprints as templates for new labeling projects.

Resources

GitHubOrganization

Sources

aid: surge-ai
name: Surge AI
description: Surge AI is a human-data company that provides large-scale, expert-quality labeled data for training and evaluating
  frontier AI models. The product surface spans RL environments and agents (rich, complex environments that challenge agentic
  models), rubrics and verifiers (scoring systems for AI outputs), RLHF (preference and reward data), SFT (foundational
  skill demonstrations), human evaluation, expert professional domains, internationalization across 70+ languages, multimodal
  (image, audio, video) data, and off-the-shelf datasets. Surge ships an official Python SDK (surge-python) wrapping the
  Surge API, with API-key authentication, and exposes the dashboard and API reference at app.surgehq.ai. Public datasets
  published by Surge include the toxicity dataset (the world's largest social-media toxicity dataset).
type: Index
image: https://kinlane-productions.s3.amazonaws.com/apis-json/apis-json-logo.jpg
tags:
- Human Data
- RLHF
- SFT
- Rubrics
- Verifiers
- RL Environments
- Multimodal
- Internationalization
- Labeling
url: https://raw.githubusercontent.com/api-evangelist/surge-ai/refs/heads/main/apis.yml
created: '2026-05-23'
modified: '2026-05-23'
specificationVersion: '0.19'
apis:
- aid: surge-ai:surge-api
  name: Surge API
  description: Surge's REST API for managing labeling projects, tasks, and results. Endpoints cover projects (list, retrieve,
    create, download results, save reports in multiple formats), tasks (create, list, retrieve individual tasks), and blueprints
    (list and use as templates for new projects). Authentication uses an API key sourced from the user's Surge profile, passed
    via the SURGE_API_KEY environment variable or set explicitly on the client. The reference is published in the Surge dashboard
    at app.surgehq.ai/docs/api.
  humanURL: https://app.surgehq.ai/docs/api
  tags:
  - REST API
  - Projects
  - Tasks
  - Blueprints
  properties:
  - type: Documentation
    url: https://app.surgehq.ai/docs/api
  - type: APIReference
    url: https://app.surgehq.ai/docs/api
  - type: Authentication
    url: https://app.surgehq.ai/docs/api
  - type: SDK
    url: https://github.com/surge-ai/surge-python
- aid: surge-ai:surge-python-sdk
  name: Surge Python SDK
  description: Official Python SDK (surge-api on PyPI) wrapping the Surge API. Requires Python 3.10+, MIT-licensed, and last
    updated May 2026. Configured via surge.api_key or the SURGE_API_KEY environment variable.
  humanURL: https://github.com/surge-ai/surge-python
  tags:
  - SDK
  - Python
  - Open Source
  properties:
  - type: GitHubRepository
    url: https://github.com/surge-ai/surge-python
  - type: SDK
    url: https://pypi.org/project/surge-api/
- aid: surge-ai:surge-rl-environments
  name: Surge RL Environments and Agents
  description: Surge's product surface for delivering complex reinforcement-learning environments and agents that challenge
    and evaluate agentic models.
  humanURL: https://www.surgehq.ai/products
  tags:
  - RL Environments
  - Agents
  - Evals
  properties:
  - type: Documentation
    url: https://www.surgehq.ai/products
- aid: surge-ai:surge-rubrics-verifiers
  name: Surge Rubrics and Verifiers
  description: Scoring rubrics and automated verifiers for grading AI outputs across domains.
  humanURL: https://www.surgehq.ai/products
  tags:
  - Rubrics
  - Verifiers
  - Evals
  properties:
  - type: Documentation
    url: https://www.surgehq.ai/products
- aid: surge-ai:surge-rlhf
  name: Surge RLHF
  description: Preference and reward data for reinforcement learning from human feedback.
  humanURL: https://www.surgehq.ai/products
  tags:
  - RLHF
  - Preference Data
  properties:
  - type: Documentation
    url: https://www.surgehq.ai/products
- aid: surge-ai:surge-sft
  name: Surge SFT
  description: Foundational-skill demonstration data for supervised fine-tuning.
  humanURL: https://www.surgehq.ai/products
  tags:
  - SFT
  - Fine-Tuning
  properties:
  - type: Documentation
    url: https://www.surgehq.ai/products
- aid: surge-ai:surge-human-evaluation
  name: Surge Human Evaluation
  description: Quality assessment of AI outputs by Surge's expert workforce.
  humanURL: https://www.surgehq.ai/products
  tags:
  - Human Evaluation
  - Quality
  properties:
  - type: Documentation
    url: https://www.surgehq.ai/products
- aid: surge-ai:surge-multimodal
  name: Surge Multimodal Data
  description: Image, audio, and video data collection and labeling.
  humanURL: https://www.surgehq.ai/products
  tags:
  - Multimodal
  - Image
  - Audio
  - Video
  properties:
  - type: Documentation
    url: https://www.surgehq.ai/products
- aid: surge-ai:surge-internationalization
  name: Surge Internationalization
  description: Multilingual data across 70+ languages for localization, translation, and multilingual model evaluation.
  humanURL: https://www.surgehq.ai/products
  tags:
  - Internationalization
  - Multilingual
  - Translation
  properties:
  - type: Documentation
    url: https://www.surgehq.ai/products
- aid: surge-ai:surge-off-the-shelf-data
  name: Surge Off-The-Shelf Data
  description: Pre-built datasets ready for licensing and download.
  humanURL: https://www.surgehq.ai/products
  tags:
  - Datasets
  - Pre-Built Data
  properties:
  - type: Documentation
    url: https://www.surgehq.ai/products
- aid: surge-ai:surge-toxicity-dataset
  name: Surge Toxicity Dataset
  description: The world's largest open social-media toxicity dataset, published under MIT license.
  humanURL: https://github.com/surge-ai/toxicity
  tags:
  - Dataset
  - Open Data
  - Toxicity
  - Trust and Safety
  properties:
  - type: GitHubRepository
    url: https://github.com/surge-ai/toxicity
common:
- type: Portal
  url: https://www.surgehq.ai
- type: Documentation
  url: https://app.surgehq.ai/docs/api
- type: APIReference
  url: https://app.surgehq.ai/docs/api
- type: Authentication
  url: https://app.surgehq.ai/docs/api
- type: SignUp
  url: https://app.surgehq.ai/customers/sign_in
- type: Console
  url: https://app.surgehq.ai
- type: SDK
  url: https://github.com/surge-ai/surge-python
  name: Surge Python SDK
- type: SDK
  url: https://pypi.org/project/surge-api/
  name: surge-api on PyPI
- type: GitHubOrganization
  url: https://github.com/surge-ai
- type: GitHubRepository
  url: https://github.com/surge-ai/toxicity
  name: Surge Toxicity Dataset
- type: Blog
  url: https://www.surgehq.ai/blog
- type: Support
  url: https://www.surgehq.ai
- type: X
  url: https://x.com/HelloSurgeAI
- type: Features
  data:
  - name: Surge REST API
    description: Endpoints for projects, tasks, and blueprints, with API-key authentication.
  - name: Python SDK
    description: Official surge-python SDK on PyPI, MIT-licensed, Python 3.10+.
  - name: RL Environments and Agents
    description: Complex environments that challenge agentic models.
  - name: Rubrics and Verifiers
    description: Scoring systems for AI outputs across domains.
  - name: RLHF and SFT
    description: Preference, reward, and demonstration data for foundation-model training.
  - name: Human Evaluation
    description: Expert workforce grades AI output quality.
  - name: Expert Professional Domains
    description: Specialized expertise across finance, law, medicine, and more.
  - name: 70+ Languages
    description: Internationalization coverage spanning more than 70 languages.
  - name: Multimodal Data
    description: Image, audio, and video collection and labeling.
  - name: Off-The-Shelf Datasets
    description: Pre-built datasets available for licensing.
  - name: Open Datasets
    description: Public releases including the world's largest social-media toxicity dataset.
- type: UseCases
  data:
  - name: Frontier Model RLHF
    description: Preference and reward data for reinforcement learning from human feedback.
  - name: Supervised Fine-Tuning
    description: Demonstration data for SFT across professional domains.
  - name: Agentic Evals
    description: Benchmark agents in complex RL environments with structured rubrics.
  - name: Multilingual Model Evaluation
    description: Evaluate model quality across 70+ languages.
  - name: Trust and Safety Research
    description: Use the toxicity dataset and human evaluation pipelines for trust and safety work.
- type: Integrations
  data:
  - name: Python SDK
    description: Programmatic integration via the official surge-python SDK.
  - name: API Key Authentication
    description: Standard API-key auth (SURGE_API_KEY env var or programmatic configuration).
  - name: Custom Project Blueprints
    description: Use Surge blueprints as templates for new labeling projects.
maintainers:
- FN: Kin Lane
  url: http://apievangelist.com
  email: [email protected]