> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.nvidia.com/nemo/gym/v0.7.0/build-verifiers/verification-patterns/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.nvidia.com/nemo/gym/_mcp/server.

# Verification Patterns

> Patterns for evaluating agent behavior and computing scores.

Verification is how an environment evaluates agent behavior and computes a score. Every environment implements some form of verification — the pattern you choose depends on your task.

#### [Equivalence Match](/build-verifiers/verification-patterns/equivalence-match)

Compare the agent's output to a known reference answer.

#### [Execution and State Match](/build-verifiers/verification-patterns/execution-and-state-match)

Execute the agent's actions, such as tool calls or generated code, and verify the output or resulting state.

#### [LLM-as-Judge](/build-verifiers/verification-patterns/llm-as-judge)

Prompt an LLM to evaluate the agent's output against rubrics, instructions, or reference answers.

#### Reward Model

Use an LLM trained on human preferences to score outputs for alignment, such as RLHF reward modeling.

coming soon

> Patterns for evaluating agent behavior and computing scores.

## Docs

- [Equivalence Match](https://docs.nvidia.com/nemo/gym/v0.7.0/build-verifiers/verification-patterns/equivalence-match.md): Extract an agent's answer and compare it deterministically with a reference value, structure, or rule.
- [Execution and State Match](https://docs.nvidia.com/nemo/gym/v0.7.0/build-verifiers/verification-patterns/execution-and-state-match.md): Run generated code, queries, actions, or proofs and score the observed output or resulting state.
- [LLM-as-Judge](https://docs.nvidia.com/nemo/gym/v0.7.0/build-verifiers/verification-patterns/llm-as-judge.md): Prompt an LLM to evaluate the agent's output against rubrics, instructions, or reference answers.