# Evaluation

> Free resources, with credit to the original creators.

## Resources
- [AI Evals FAQ](/evaluation/application-evals/hamel-husain-ai-evals-faq/index.md): Practical answers about error analysis, annotation and reliable evaluators. (intermediate; West; closed resource; by Hamel Husain and Shreya Shankar)
- [Your AI Product Needs Evals](/evaluation/application-evals/hamel-husain-your-ai-product-needs-evals/index.md): Build a feedback loop from real failures, not general benchmark scores. (intermediate; West; closed resource; by Hamel Husain)
- [SWE-bench Leaderboards](/evaluation/agent-evals/swe-bench-team-swe-bench-leaderboards/index.md): Understand coding-agent scores in terms of resolved repository tasks. (intermediate; West; closed resource; by SWE-bench team (Princeton/Stanford))
- [lm-evaluation-harness](/evaluation/eval-harnesses/eleutherai-lm-evaluation-harness/index.md): Run model benchmarks through reusable task and backend interfaces. (advanced; West; open resource; by EleutherAI)
- [Evaluating the Effectiveness of LLM-Evaluators (LLM-as-Judge)](/evaluation/llm-as-judge/eugene-yan-evaluating-the-effectiveness-of-llm-evaluators/index.md): Compare model-based graders against the judgments they must approximate. (advanced; West; closed resource; by Eugene Yan)
- [simple-evals](/evaluation/eval-harnesses/openai-simple-evals/index.md): Read reference evaluation code with an explicit maintenance caveat. (advanced; West; open resource; by OpenAI)
- [Inspect](/evaluation/eval-harnesses/uk-ai-security-institute-inspect/index.md): Compose evaluation datasets, agents, tools and scorers. (advanced; West; closed resource; by UK AI Security Institute and Meridian Labs)