AI Tool
Giskard pricing, features, company info, and alternatives
A factual product page for Giskard's AI red teaming and LLM evaluation platform.
Last updated April 2026 · Pricing and features verified against official documentation
Pricing
Current public pricing tiers on file for Giskard, last verified Apr 25, 2026.
Open Source
$0
Open-source library for local use, basic LLM vulnerability scans, and basic RAG evaluation workflows.
Enterprise
Custom
Book a demo; includes continuous red teaming, AI agent quality evaluation, collaboration, RBAC, audit trails, SSO, hybrid deployment, data residency, 0-training policy, and SLAs.
What You Can Do With It
The main capabilities that shape how people use Giskard today.
Combines continuous red teaming, LLM evaluation, and an open-source Python library in one product family.
Supports shared workspaces, dataset versioning, custom failure categories, alerting, and cron-scheduled evaluations in the Hub.
Exposes the Hub through a Python SDK and API, while the open-source library covers LLM Scan and RAG Evaluation Toolkit workflows.
Offers hybrid deployment options, data residency controls, and enterprise security features including SSO, RBAC, audit logs, and 0-training policy.
Best For
Who Giskard is most clearly built for.
Teams that need continuous red teaming and evaluation for production LLM agents.
Enterprise buyers that want RBAC, audit trails, SSO, and hybrid deployment options.
Developers who want a Python SDK and an open-source library alongside the Hub.
Company
Leadership and company context for Giskard AI SAS.
CEO
Alex Combessie and Jean-Marie John-Mathews (co-CEOs)
Founders
Alex Combessie, Jean-Marie John-Mathews
Headquarters
13 rue des Petits Hotels, 75010 Paris, France
Platforms
Where you can use Giskard today.
Web
Python SDK
API
Open-source Python library
SaaS
Private cloud
On-premises
Integrations
Notable connected tools and ecosystem hooks for Giskard.
GitHub
MLflow
Weights & Biases
NeMo Guardrails
Hugging Face
Pytest
Privacy Notes
Publicly stated data-handling notes that matter when evaluating Giskard.
The homepage says Giskard can process data in the EU or US with data residency and isolation controls.
The homepage and pricing page say the platform uses end-to-end encryption at rest and in transit and a 0-training policy.
The privacy policy says Giskard AI is based in Paris, France and processes website and product navigation data under GDPR terms.
Compliance
Public compliance or enterprise-governance signals we found for Giskard.
GDPR
SOC 2 Type II
HIPAA
Access
How to integrate or build around Giskard.
Public API
Yes
Docs
Available
Alternatives
Other tools worth considering alongside Giskard.
Framework-agnostic platform for observability, evaluation, and deployment of AI agents and LLM apps.
Open-source LLM engineering platform for tracing, prompt management, evaluations, and analytics.
AI observability and evaluation platform for tracing, scoring, and improving production AI applications.
Open-source AI observability and evaluation platform for tracing, prompt iteration, and experiments.
Product Snapshot
Giskard is an AI agent evaluation and red teaming platform. Its documentation distinguishes between the enterprise Hub and the open-source Python library.
What You Can Do With It
- Run continuous red teaming and vulnerability scans against LLM agents.
- Build and version evaluation datasets, checks, and scheduled evaluations in the Hub.
- Use the Python SDK and API for programmatic workflows.
- Use the open-source library for LLM Scan and RAG Evaluation Toolkit workflows.
Why It Stands Out
It combines an enterprise Hub, an open-source library, and research-backed safety work in one product line.
Tradeoffs To Know
- The enterprise plan is quote-based, so you need to book a demo for pricing details.
- The free tier is aimed at solo experiments rather than production deployment.
- The strongest governance features sit in the enterprise Hub, not the open-source library.
Sources
- giskard.ai/pricing
- giskard.ai/legal
- notion.giskard.ai/Giskard-Privacy-Policy-24fdf583d5c980a78942f984e23802b5
- giskard.ai/about
- giskard.ai
- docs.giskard.ai
- docs.giskard.ai/hub/sdk/reference
- docs.giskard.ai/en/latest/integrations/cicd/index.html
- docs.giskard.ai/en/latest/integrations/mlflow/index.html
- docs.giskard.ai/en/stable/integrations/wandb/index.html
- docs.giskard.ai/en/latest/integrations/nemoguardrails/index.html
- docs.giskard.ai/en/stable/integrations/huggingface/index.html
- docs.giskard.ai/en/stable/integrations/pytest/index.html
- trust.giskard.ai