Home » AI Directory » Developer » DeepEval
DeepEval
Open-source framework for evaluating LLM outputs.
Visit DeepEval » github.com/confident-ai/deepeval
| DeepEval at a glance | |
|---|---|
| What it does | Open-source framework for evaluating LLM outputs |
| Category | Developer |
| Website | github.com/confident-ai/deepeval |
| Checked | Website working when last checked (2026-10-16) |
In their own words
“The LLM Evaluation Framework. Contribute to confident-ai/deepeval development by creating an account on GitHub.”
— from the DeepEval website
8 alternatives to DeepEval
Other developer tools that do similar things:
Guardrails AI
Open-source framework for validating LLM outputs
Haystack
Open-source framework for LLM and RAG pipelines
Vanna
Open-source text-to-SQL framework using RAG
AutoGen
Microsoft open-source framework for multi-agent AI apps
LiteLLM
Open-source LLM gateway with a unified API
Helicone
Open-source LLM observability and gateway
Langfuse
Open-source LLM engineering platform for tracing and evals
Arize Phoenix
Open-source LLM tracing and evaluation
How to choose
- Try it on your real codebase, not a demo. Results vary a lot by language and project size.
- Decide how much control you want: suggestions you accept line by line, or an agent that makes larger changes for you to review.
- Check your company's rules on sending code to outside AI services before you start.
Full guide to choosing developer tools »
Is this your tool? Get it featured at the top of its category. Something wrong on this page? Tell us.