tools
Palantir AIP
Enterprise AI platform for connecting large language models to data, workflows, applications, and operational processes.
Palantir AIP connects large language models to an organization’s data and operations. It provides tools for building AI workflows, agents, applications, functions, and automations on top of Palantir Foundry and its Ontology.
Organizations use AIP for governed AI development, operational decision support, analytics, and process automation. It supports multiple model providers and custom models, but availability varies by enrollment. Usage of Palantir-provided models incurs compute costs, and enterprise pricing depends on contract terms.
Features
- Build LLM-backed workflows, agents, and applications
- Create functions with AIP Logic
- Develop chatbots with AIP Chatbot Studio
- Connect models to Foundry Ontology data, logic, and actions
- Use AIP Analyst to explore ontology data and generate analyses
- Evaluate LLM workflows with evaluation suites
- Connect custom or self-hosted models through registered models
- Use Python, Java, and TypeScript SDKs for AIP-powered applications
Use cases
- Build AI assistants grounded in enterprise data
- Automate operational workflows with agents and functions
- Analyze ontology data with natural-language questions
- Create governed customer-service and supply-chain applications
- Evaluate and improve production LLM workflows
- Deploy models across enterprise, government, and defense environments
Pros
Cons
Latest updates
- DeepSeek V4.1 Flash (1017) support in AIP through Fireworks
AIP adds experimental support for DeepSeek V4.1 Flash (1017) through Fireworks on non-georestricted commercial enrollments.
- Bring public web information into your AIP workflows
Web search lets supported OpenAI models in AIP find information from public web sources and return answers with citations.
- Pipeline builder defaults to faster pipelines, now with LLM support
Faster pipelines are now the default in Pipeline Builder and support Use LLM nodes and Text to Embeddings nodes.
- Foundry Worker data ingest and export, media ingest, and Iceberg extract operations are available without compute costs
Data ingest, data export, media ingest, and Iceberg extract operations on Foundry Worker no longer incur compute costs.
- GLM-5.3 now available in AIP through Amazon Bedrock
GLM-5.3 is available as an experimental model in AIP through Amazon Bedrock.
Capabilities
- Reads your files — “just drag and drop documents into AIP Threads” source
- Voice conversation — “Build a voice-enabled OSDK application” source
- Acts for you — “AIP Chatbot Studio enables you to create interactive chatbots that can leverage enterprise-specific data in the Ontology and a range of tools to complete tasks and achieve goals.” source
- Developer API — “Use AIP Chatbots through Foundry APIs” source
- Mobile app — “AIP forms an operating system that can deliver a full range of AI-driven products, from LLM-powered web applications to mobile applications using vision-language models to edge applications that embed localized AI.” source
Get it
Pricing
- Starting price
- Free
- Prices checked
- 2026-10-26
- Read
- by web search
Developer Tier
Free
- Free limited developer access to Foundry/AIP
- Limited capacity for development and evaluation
- Not for commercial/production deployment
Commercial / Enterprise
Price on request
- Negotiated annual platform fee
- Scoped by users, data domains, compute, and use cases
- Usage-based LLM compute (compute-seconds, rates vary by model/region; contact representative for contract terms)