AI Tools
DeepEval
DeepEval: Preise, Fähigkeiten, Updates und Alternativen, von den Seiten des Anbieters gelesen.

Auf Englisch
Key facts
- Price
- Free plan · paid plans: see pricing Read on deepeval.com
- Free plan
- Yes Read on deepeval.com
- Platforms
- Web · CLI · API Read on confident-ai.com
- API
- Yes
- Open source
- Yes · Apache-2.0 Read on github.com
- Latest version
- v0.9.26 Read on deepeval.com
- Last update
- 11 Nov 2026 Read on github.com
- Prices checked
- 27 Oct 2026 Read on deepeval.com
DeepEval lets developers write tests for LLM outputs, run evaluations on agents, chatbots, RAG and other AI workflows, and generate synthetic test data. It supports Python and TypeScript workflows, including pytest-style tests and CI/CD checks.
The framework is free and Apache 2.0 licensed. Confident AI is a separate, optional cloud platform for team collaboration, tracing and other hosted features; its plans are listed below. Some evaluation metrics use an external LLM provider, which requires a provider key.
- Category
- Developer Tools
- Read from the maker’s pages
- 2026-11-18
Company
Features
- Test LLM outputs with pytest-style assertions
- Evaluate RAG, agents, multi-turn chat, voice and multimodal systems
- Run end-to-end, trajectory-based and component-level evaluations
- Use built-in or custom evaluation metrics
- Generate synthetic datasets and simulate conversations
- Optimize prompts and run evaluations in CI/CD
- Apache 2.0 licensed
Use cases
- Test LLM output quality before releasing a prompt or model change
- Gate AI application changes with evaluations in CI/CD
- Measure retrieval quality in a RAG pipeline
- Evaluate an agent's tool use and execution steps
- Generate test data for edge cases that are hard to collect
Pricing
Free plan · paid plans: see pricing
- Prices checked
- 2026-10-27
- Read
- by web search
DeepEval
free
- Open-source LLM evaluation framework
- Apache 2.0 licensed
- Local and CI/CD test runner
Free
free
- Full LLM unit and regression testing suite
- Evals in development and CI/CD
- LLM tracing
- Prompt versioning
- Datasets on the cloud
- Limited to 2 user seats, 1 project, 5 test runs per week, 1 GB-month of trace spans
Starter
$200/mo
- $200 / month
- Everything in Free, plus no-code AI evaluation workflows
- Custom evaluation metrics
- Online evals and classifications on live traffic
- Annotation queues & workflows
- Real-time alerting
- Unlimited user seats, 5 projects, 5 GB-months of trace spans then $1 per GB-month
Team
$2000/mo
- $2,000 / month
- Everything in Starter, plus metric & dataset versioning
- Git-based prompt workflows
- Custom RBAC, SOC2, SSO
- Dedicated support channel
- Unlimited user seats and projects, 75 GB-months of trace spans then $1 per GB-month
Enterprise
price on request
- Everything in Team, plus advanced AI authentication options
- Dedicated On-Prem Deployment, HIPAA
- Custom data residency
- Dedicated 24x7 technical support
- Unlimited user seats, projects, GB-months, online eval metric runs/month
Capabilities
- Runs commands — “Pytest-native evals that run in CI/CD or as Python scripts.” source
- Command line — “50+ metrics, one CLI” source
- Official SDKs — “Pytest-native evals that run in CI/CD or as Python scripts.” source
- Traces and evaluates — “DeepEval traces every step of your agent into something you can grade, and improve — visible in your terminal, testable in your runner, shippable in your next commit.” source
Latest updates
- typescript-v0.9.26
New release.
- python-v4.3.4
New release.
- typescript-v0.9.25
New release.
- python-v4.3.3
New release.
- python-v4.3.2
New release.
Where it works
Web · CLI · API
Alternatives to DeepEval
- Promptfoo
Promptfoo tests AI prompts and applications, compares model outputs, and scans for security vulnerabilities.
- Langfuse
Langfuse traces AI applications, manages prompts, and evaluates model and agent outputs using cloud or self-hosted deployments.
- Arize Phoenix
Arize Phoenix helps developers trace, evaluate, debug, and improve AI applications and agents.
- Braintrust
Braintrust helps teams inspect AI-agent activity, evaluate outputs, and find production issues.
- Opik
Opik traces, evaluates, and monitors LLM apps and AI agents, from development through production.
- MLflow
MLflow helps teams track machine-learning work and build, evaluate, trace, and monitor LLM applications and agents.
Questions about DeepEval
Is DeepEval free?
Yes. The maker’s pages state that DeepEval can be used for free. Read on the maker’s pages on 18 Nov 2026.
How much does DeepEval cost?
The plans on the maker’s pricing page: DeepEval (free), Free (free), Starter ($200/mo), Team ($2000/mo) and Enterprise (price on request). These prices were read by web search, not on the maker’s own page: check its pricing page before relying on them. Prices checked on 27 Oct 2026.
Does DeepEval have an API?
Yes. The maker’s pages state that DeepEval has an API. Read on the maker’s pages on 18 Nov 2026.
Which platforms does DeepEval run on?
DeepEval runs on Web, CLI and API, as its maker’s pages state. Read on the maker’s pages on 18 Nov 2026.
Is DeepEval open source?
Yes. The code of DeepEval is published at github.com/confident-ai/deepeval under the Apache-2.0 licence. Read on the maker’s pages on 18 Nov 2026.
Is DeepEval still available?
Yes, as of 18 Nov 2026: its maker’s pages were read that day and carried no notice that DeepEval has ended.
What changed recently in DeepEval?
The latest entry in its maker’s changelog: 11 Nov 2026, “typescript-v0.9.26”. Before it: 11 Nov 2026, “python-v4.3.4”. Changelog checked on 19 Nov 2026.
Sources
- https://www.confident-ai.com/pricing — Official pricing page read on 2026-11-18; lists Confident AI cloud plans and features.
- https://deepeval.com/docs/introduction — Describes DeepEval's evaluation capabilities, local-first approach and displayed Python version.
- https://deepeval.com/changelog/changelog-2026 — Lists November updates, including v4.3.4; no exact release date is shown.
- https://www.confident-ai.com/frameworks/deepeval — Confirms DeepEval is free and Apache 2.0 licensed, and distinguishes it from the paid cloud platform.
- https://deepeval.com/docs/faq — Explains provider keys, local use and the distinction between DeepEval and Confident AI.
Not confirmed
- Prices
- These are Confident AI cloud-platform plans, not paid tiers for the free DeepEval framework.