# Verifiers

## Docs

- [AnthropicMessagesClient](https://mintlify.wiki/primeintellect-ai/verifiers/api/anthropic-client.md): Anthropic Messages API client implementation
- [Client](https://mintlify.wiki/primeintellect-ai/verifiers/api/client.md): Base abstract client class for LLM providers
- [EnvGroup](https://mintlify.wiki/primeintellect-ai/verifiers/api/env-group.md): Mixture of multiple environments with task-based routing
- [Environment](https://mintlify.wiki/primeintellect-ai/verifiers/api/environment.md): Base class for all RL environments
- [CliAgentEnv](https://mintlify.wiki/primeintellect-ai/verifiers/api/experimental/cli-agent-env.md): Environment for running custom agent code in sandboxes with API interception
- [GymEnv](https://mintlify.wiki/primeintellect-ai/verifiers/api/experimental/gym-env.md): Universal runner for Gym-compatible environments
- [HarborEnv](https://mintlify.wiki/primeintellect-ai/verifiers/api/experimental/harbor-env.md): CliAgentEnv subclass for loading Harbor-format tasks
- [MCPEnv](https://mintlify.wiki/primeintellect-ai/verifiers/api/experimental/mcp-env.md): Environment for integrating MCP (Model Context Protocol) servers as tools
- [RolloutGatewayMixin](https://mintlify.wiki/primeintellect-ai/verifiers/api/experimental/rollout-gateway-mixin.md): Server-side gateway for CliAgentEnv with direct model communication
- [BrowserEnv](https://mintlify.wiki/primeintellect-ai/verifiers/api/integrations/browser-env.md): Unified browser automation with DOM and CUA modes
- [OpenEnvEnv](https://mintlify.wiki/primeintellect-ai/verifiers/api/integrations/openenv-env.md): Drop-in OpenEnv integration for Verifiers
- [ReasoningGymEnv](https://mintlify.wiki/primeintellect-ai/verifiers/api/integrations/reasoning-gym-env.md): Wrapper for Reasoning Gym procedural reasoning datasets
- [TextArenaEnv](https://mintlify.wiki/primeintellect-ai/verifiers/api/integrations/textarena-env.md): Wrapper for TextArena text-based game environments
- [JudgeRubric](https://mintlify.wiki/primeintellect-ai/verifiers/api/judge-rubric.md): LLM-as-judge scoring rubric for evaluating model responses
- [MathRubric](https://mintlify.wiki/primeintellect-ai/verifiers/api/math-rubric.md): Specialized rubric for mathematical equivalence checking
- [MaybeThinkParser](https://mintlify.wiki/primeintellect-ai/verifiers/api/maybe-think-parser.md): Parser that handles responses with or without <think>...</think> tags
- [MultiTurnEnv](https://mintlify.wiki/primeintellect-ai/verifiers/api/multi-turn-env.md): Environment for multi-turn interactive tasks
- [OpenAIChatCompletionsClient](https://mintlify.wiki/primeintellect-ai/verifiers/api/openai-client.md): OpenAI Chat Completions API client implementation
- [OpenAICompletionsClient](https://mintlify.wiki/primeintellect-ai/verifiers/api/openai-completions-client.md): OpenAI Completions API client for legacy text completion models
- [OpenAIChatCompletionsTokenClient](https://mintlify.wiki/primeintellect-ai/verifiers/api/openai-token-client.md): vLLM token-level client with TITO (token-in, token-out) optimization
- [Parser](https://mintlify.wiki/primeintellect-ai/verifiers/api/parser.md): Base class for parsing LLM completions and extracting answers
- [PythonEnv](https://mintlify.wiki/primeintellect-ai/verifiers/api/python-env.md): Sandbox environment with persistent Python REPL
- [Rubric](https://mintlify.wiki/primeintellect-ai/verifiers/api/rubric.md): Base class for reward functions in RL environments
- [RubricGroup](https://mintlify.wiki/primeintellect-ai/verifiers/api/rubric-group.md): Aggregate multiple rubrics to evaluate rollouts with combined reward functions
- [SandboxEnv](https://mintlify.wiki/primeintellect-ai/verifiers/api/sandbox-env.md): Sandboxed container execution environment
- [SingleTurnEnv](https://mintlify.wiki/primeintellect-ai/verifiers/api/single-turn-env.md): Environment for single-turn Q&A tasks
- [StatefulToolEnv](https://mintlify.wiki/primeintellect-ai/verifiers/api/stateful-tool-env.md): Environment for tool-calling tasks with per-rollout state
- [ThinkParser](https://mintlify.wiki/primeintellect-ai/verifiers/api/think-parser.md): Parser for models that include <think>...</think> tags but do not parse them automatically
- [ToolEnv](https://mintlify.wiki/primeintellect-ai/verifiers/api/tool-env.md): Environment for stateless tool-calling tasks
- [Messages](https://mintlify.wiki/primeintellect-ai/verifiers/api/types/messages.md): Message types for chat-based interactions
- [Reward Functions](https://mintlify.wiki/primeintellect-ai/verifiers/api/types/reward-func.md): Types for reward and metric functions
- [State](https://mintlify.wiki/primeintellect-ai/verifiers/api/types/state.md): Runtime state dictionary for rollouts
- [Config Utilities](https://mintlify.wiki/primeintellect-ai/verifiers/api/utils/config.md): Environment variable validation utilities
- [Data Utilities](https://mintlify.wiki/primeintellect-ai/verifiers/api/utils/data.md): Helper functions for dataset processing
- [Logging Utilities](https://mintlify.wiki/primeintellect-ai/verifiers/api/utils/logging.md): Logging configuration and utilities for Verifiers
- [XMLParser](https://mintlify.wiki/primeintellect-ai/verifiers/api/xml-parser.md): Parser for extracting structured data from XML-formatted completions
- [prime env init](https://mintlify.wiki/primeintellect-ai/verifiers/cli/env-init.md): Create a new environment template
- [prime env install](https://mintlify.wiki/primeintellect-ai/verifiers/cli/env-install.md): Install an environment package
- [prime env push](https://mintlify.wiki/primeintellect-ai/verifiers/cli/env-push.md): Publish an environment to the Environments Hub
- [prime eval run](https://mintlify.wiki/primeintellect-ai/verifiers/cli/eval-run.md): Run evaluations on environments with any model
- [prime eval tui](https://mintlify.wiki/primeintellect-ai/verifiers/cli/eval-tui.md): View evaluation results in an interactive terminal UI
- [prime lab setup](https://mintlify.wiki/primeintellect-ai/verifiers/cli/lab-setup.md): Initialize a development workspace for building RL environments
- [CLI Overview](https://mintlify.wiki/primeintellect-ai/verifiers/cli/overview.md): Command-line interface for the Verifiers SDK
- [Legacy vf-* Commands](https://mintlify.wiki/primeintellect-ai/verifiers/cli/vf-commands.md): Backward-compatible CLI commands from earlier verifiers versions
- [Datasets](https://mintlify.wiki/primeintellect-ai/verifiers/concepts/datasets.md): How datasets work with HuggingFace integration and lazy loading
- [Environments](https://mintlify.wiki/primeintellect-ai/verifiers/concepts/environments.md): Understanding environment types and how they orchestrate RL training and evaluation
- [Rollouts](https://mintlify.wiki/primeintellect-ai/verifiers/concepts/rollouts.md): The rollout loop, state management, and trajectory tracking
- [Rubrics](https://mintlify.wiki/primeintellect-ai/verifiers/concepts/rubrics.md): Reward functions and scoring system for RL training
- [Browser Automation Examples](https://mintlify.wiki/primeintellect-ai/verifiers/examples/browser-examples.md): Control web browsers with DOM manipulation and vision-based interactions
- [GSM8K Math Reasoning](https://mintlify.wiki/primeintellect-ai/verifiers/examples/gsm8k.md): Grade school math problems with step-by-step reasoning
- [Math with Python Execution](https://mintlify.wiki/primeintellect-ai/verifiers/examples/math-python.md): Solve mathematical problems with Python code execution in sandboxes
- [Wikipedia Search Tool](https://mintlify.wiki/primeintellect-ai/verifiers/examples/wiki-search.md): Answer questions using Wikipedia search tools with embeddings and multi-turn interaction
- [Wordle Game Environment](https://mintlify.wiki/primeintellect-ai/verifiers/examples/wordle.md): Train models to play Wordle using multi-turn interaction and feedback
- [Custom Multi-Turn Patterns](https://mintlify.wiki/primeintellect-ai/verifiers/guides/custom-environments.md): Advanced environment customization for complex interaction patterns
- [Running Evaluations](https://mintlify.wiki/primeintellect-ai/verifiers/guides/evaluation.md): Evaluate environments with models and analyze results
- [Building Multi-Turn Environments](https://mintlify.wiki/primeintellect-ai/verifiers/guides/multi-turn.md): Create interactive environments with turn-by-turn model responses
- [Building Single-Turn Environments](https://mintlify.wiki/primeintellect-ai/verifiers/guides/single-turn.md): Create simple question-answering environments with the Verifiers SDK
- [Tool-Use Environments](https://mintlify.wiki/primeintellect-ai/verifiers/guides/tool-environments.md): Build environments where models can call tools and execute code
- [RL Training Guide](https://mintlify.wiki/primeintellect-ai/verifiers/guides/training.md): Train models with your Verifiers environments using RL
- [Installation](https://mintlify.wiki/primeintellect-ai/verifiers/installation.md): Install Verifiers and set up your development environment
- [Environments Hub](https://mintlify.wiki/primeintellect-ai/verifiers/integrations/environments-hub.md): Publish and discover Verifiers environments
- [Hosted Training](https://mintlify.wiki/primeintellect-ai/verifiers/integrations/hosted-training.md): Train models on Prime Intellect's managed infrastructure
- [MCP Integration](https://mintlify.wiki/primeintellect-ai/verifiers/integrations/mcp.md): Integrate Model Context Protocol servers as tools in Verifiers environments
- [OpenEnv Integration](https://mintlify.wiki/primeintellect-ai/verifiers/integrations/openenv.md): Wrap OpenEnv environments for use with Verifiers
- [Prime-RL Integration](https://mintlify.wiki/primeintellect-ai/verifiers/integrations/prime-rl.md): Train models with the prime-rl framework using Verifiers environments
- [Reasoning Gym Integration](https://mintlify.wiki/primeintellect-ai/verifiers/integrations/reasoning-gym.md): Use Reasoning Gym procedural datasets in Verifiers environments
- [TextArena Integration](https://mintlify.wiki/primeintellect-ai/verifiers/integrations/textarena.md): Wrap TextArena game environments for use with Verifiers
- [Verifiers](https://mintlify.wiki/primeintellect-ai/verifiers/introduction.md): Environments for LLM Reinforcement Learning
- [Quick Start](https://mintlify.wiki/primeintellect-ai/verifiers/quickstart.md): Get started with Verifiers in minutes