Document Search Engine
Context Graph / DSL for Contract Semantics / Agentic RAG
Pre-computed legal knowledge and Context Graphs connect contracts, amendments and related documents, enabling users and AI agents to retrieve the right evidence faster and analyse transactions in context.
Document Search Engine
Context Graph / DSL for Contract Semantics / Agentic RAG
Pre-computed legal knowledge and Context Graphs connect contracts, amendments and related documents, enabling users and AI agents to retrieve the right evidence faster and analyse transactions in context.
Document Search Engine
Context Graph / DSL for Contract Semantics / Agentic RAG
Pre-computed legal knowledge and Context Graphs connect contracts, amendments and related documents, enabling users and AI agents to retrieve the right evidence faster and analyse transactions in context.
77K
Legal transactions mapped into Context Graphs
99.4%
accurate retrieval across 1,300+ legal concepts
40K
Tested across 40,000 samples and run daily to test for regression
77K
Legal transactions mapped into Context Graphs
99.4%
accurate retrieval across 1,300+ legal concepts
40K
Tested across 40,000 samples and run daily to test for regression
77K
Legal transactions mapped into Context Graphs
99.4%
accurate retrieval across 1,300+ legal concepts
40K
Tested across 40,000 samples and run daily to test for regression
Document Search Engine Context Graph
The answer is rarely in one contract.
Finding an answer in one contract is one thing. Understanding what that contract means across the business is another.
The answer may sit in a single sentence, across multiple clauses, in an amendment that changed the original agreement, or across a family of documents that together form one legal transaction.
Many AI tools begin understanding your contracts only when you ask a question. They retrieve text, pass it to a language model and ask it to determine what matters in the moment. That can work for simple questions. At enterprise scale, every request has to rediscover the relevant legal knowledge, precedent and document relationships before it can reason over them.
Luminance does that work upfront. It pre-computes the legal knowledge within every contract and connects related agreements, amendments and transaction context. So teams can make sense of the full picture in seconds, surface what matters and act with confidence.
Document Search Engine Context Graph
The answer is rarely in one contract.
Finding an answer in one contract is one thing. Understanding what that contract means across the business is another.
The answer may sit in a single sentence, across multiple clauses, in an amendment that changed the original agreement, or across a family of documents that together form one legal transaction.
Many AI tools begin understanding your contracts only when you ask a question. They retrieve text, pass it to a language model and ask it to determine what matters in the moment. That can work for simple questions. At enterprise scale, every request has to rediscover the relevant legal knowledge, precedent and document relationships before it can reason over them.
Luminance does that work upfront. It pre-computes the legal knowledge within every contract and connects related agreements, amendments and transaction context. So teams can make sense of the full picture in seconds, surface what matters and act with confidence.
Document Search Engine Context Graph
The answer is rarely in one contract.
Finding an answer in one contract is one thing. Understanding what that contract means across the business is another.
The answer may sit in a single sentence, across multiple clauses, in an amendment that changed the original agreement, or across a family of documents that together form one legal transaction.
Many AI tools begin understanding your contracts only when you ask a question. They retrieve text, pass it to a language model and ask it to determine what matters in the moment. That can work for simple questions. At enterprise scale, every request has to rediscover the relevant legal knowledge, precedent and document relationships before it can reason over them.
Luminance does that work upfront. It pre-computes the legal knowledge within every contract and connects related agreements, amendments and transaction context. So teams can make sense of the full picture in seconds, surface what matters and act with confidence.
STRUCTURE LEGAL KNOWLEDGE
Pre-compute what is known.
As documents enter the Platform, Luminance analyzes and structures their legal content upfront – identifying more than 1,300 legal concepts, as well as clauses, entities, metadata, and relationships between documents. This creates a pre-computed layer of legal knowledge that can be queried instantly when needed.
When a question is asked, Luminance doesn't have to start with raw documents. It can combine structured legal knowledge, semantic search, and the wider context of the transaction to retrieve the right evidence for the task.
Unstructured contract language becomes structured legal knowledge that users and AI agents can query through:
Structured legal knowledge: Known legal concepts, annotations, entities and document metadata are indexed, so precise questions can be answered through structured retrieval rather than regenerated from scratch.
Semantic search: Find relevant language even where the user's question does not exactly match the wording used in the contract – supporting "needle in a haystack" questions across documents.
Hybrid retrieval: Structured annotations and metadata can be combined with semantic search, using what Luminance already knows about the contracts to narrow the search and retrieve more relevant evidence.
This means the AI doesn’t have to rediscover what Luminance already knows. Instead, it can reason over the right context from the start, rather than rediscovering legal knowledge, precedent, and document relationships for every request.
STRUCTURE LEGAL KNOWLEDGE
Pre-compute what is known.
As documents enter the Platform, Luminance analyzes and structures their legal content upfront – identifying more than 1,300 legal concepts, as well as clauses, entities, metadata, and relationships between documents. This creates a pre-computed layer of legal knowledge that can be queried instantly when needed.
When a question is asked, Luminance doesn't have to start with raw documents. It can combine structured legal knowledge, semantic search, and the wider context of the transaction to retrieve the right evidence for the task.
Unstructured contract language becomes structured legal knowledge that users and AI agents can query through:
Structured legal knowledge: Known legal concepts, annotations, entities and document metadata are indexed, so precise questions can be answered through structured retrieval rather than regenerated from scratch.
Semantic search: Find relevant language even where the user's question does not exactly match the wording used in the contract – supporting "needle in a haystack" questions across documents.
Hybrid retrieval: Structured annotations and metadata can be combined with semantic search, using what Luminance already knows about the contracts to narrow the search and retrieve more relevant evidence.
This means the AI doesn’t have to rediscover what Luminance already knows. Instead, it can reason over the right context from the start, rather than rediscovering legal knowledge, precedent, and document relationships for every request.
STRUCTURE LEGAL KNOWLEDGE
Pre-compute what is known.
As documents enter the Platform, Luminance analyzes and structures their legal content upfront – identifying more than 1,300 legal concepts, as well as clauses, entities, metadata, and relationships between documents. This creates a pre-computed layer of legal knowledge that can be queried instantly when needed.
When a question is asked, Luminance doesn't have to start with raw documents. It can combine structured legal knowledge, semantic search, and the wider context of the transaction to retrieve the right evidence for the task.
Unstructured contract language becomes structured legal knowledge that users and AI agents can query through:
Structured legal knowledge: Known legal concepts, annotations, entities and document metadata are indexed, so precise questions can be answered through structured retrieval rather than regenerated from scratch.
Semantic search: Find relevant language even where the user's question does not exactly match the wording used in the contract – supporting "needle in a haystack" questions across documents.
Hybrid retrieval: Structured annotations and metadata can be combined with semantic search, using what Luminance already knows about the contracts to narrow the search and retrieve more relevant evidence.
This means the AI doesn’t have to rediscover what Luminance already knows. Instead, it can reason over the right context from the start, rather than rediscovering legal knowledge, precedent, and document relationships for every request.
THE RETRIEVAL ENGINE, BUILT FOR DIFFERENT QUESTIONS
Retrieve what matters.
Not every legal question requires the same amount of information or attention. Luminance adjusts its retrieval strategy according to the breadth and complexity of the request:
A specific fact in a small number of documents: Vector search locates relevant passages in seconds.
A specific issue across many documents: Relevant passages are retrieved across a wider document population and analyzed together.
A complex question requiring substantial context: Larger amounts of relevant material can be retrieved for deeper analysis.
Portfolio-scale analysis: Where the question requires many pieces of information across many documents, deeper analysis can move beyond real-time chat into purpose-built reporting workflows.
THE RETRIEVAL ENGINE, BUILT FOR DIFFERENT QUESTIONS
Retrieve what matters.
Not every legal question requires the same amount of information or attention. Luminance adjusts its retrieval strategy according to the breadth and complexity of the request:
A specific fact in a small number of documents: Vector search locates relevant passages in seconds.
A specific issue across many documents: Relevant passages are retrieved across a wider document population and analyzed together.
A complex question requiring substantial context: Larger amounts of relevant material can be retrieved for deeper analysis.
Portfolio-scale analysis: Where the question requires many pieces of information across many documents, deeper analysis can move beyond real-time chat into purpose-built reporting workflows.
THE RETRIEVAL ENGINE, BUILT FOR DIFFERENT QUESTIONS
Retrieve what matters.
Not every legal question requires the same amount of information or attention. Luminance adjusts its retrieval strategy according to the breadth and complexity of the request:
A specific fact in a small number of documents: Vector search locates relevant passages in seconds.
A specific issue across many documents: Relevant passages are retrieved across a wider document population and analyzed together.
A complex question requiring substantial context: Larger amounts of relevant material can be retrieved for deeper analysis.
Portfolio-scale analysis: Where the question requires many pieces of information across many documents, deeper analysis can move beyond real-time chat into purpose-built reporting workflows.
Luminance query language
Receive precise answers.
Behind the retrieval layer sits the proprietary Luminance Query Language. It gives users and AI agents a consistent way to query the structured legal knowledge stored in the platform, rather than relying entirely on natural-language search.
The language operates across:
22 field types
11 predicates and operators
100+ atomic query structures
1,300+ legal concepts
5,000+ concrete primitive expressions
64 aggregation queries
The result: Users and AI agents query the same underlying legal data from a consistent foundation.
Luminance query language
Receive precise answers.
Behind the retrieval layer sits the proprietary Luminance Query Language. It gives users and AI agents a consistent way to query the structured legal knowledge stored in the platform, rather than relying entirely on natural-language search.
The language operates across:
22 field types
11 predicates and operators
100+ atomic query structures
1,300+ legal concepts
5,000+ concrete primitive expressions
64 aggregation queries
The result: Users and AI agents query the same underlying legal data from a consistent foundation.
Luminance query language
Receive precise answers.
Behind the retrieval layer sits the proprietary Luminance Query Language. It gives users and AI agents a consistent way to query the structured legal knowledge stored in the platform, rather than relying entirely on natural-language search.
The language operates across:
22 field types
11 predicates and operators
100+ atomic query structures
1,300+ legal concepts
5,000+ concrete primitive expressions
64 aggregation queries
The result: Users and AI agents query the same underlying legal data from a consistent foundation.
See the full transaction behind every contract: from individual contracts to complete legal transactions.
An isolated contract rarely tells the whole story: an amendment may overwrite an original provision. A related agreement may change an obligation. A playbook may define the organization's preferred position. Previous negotiations may provide important commercial context.
Luminance analyzes relationships between documents and connects them into Context Graphs representing the wider legal transaction. Instead of reasoning over an isolated file, the platform can retrieve the documents and relationships surrounding it.
The contract provides the text. The Context Graph provides a bigger picture.
See the full transaction behind every contract: from individual contracts to complete legal transactions.
An isolated contract rarely tells the whole story: an amendment may overwrite an original provision. A related agreement may change an obligation. A playbook may define the organization's preferred position. Previous negotiations may provide important commercial context.
Luminance analyzes relationships between documents and connects them into Context Graphs representing the wider legal transaction. Instead of reasoning over an isolated file, the platform can retrieve the documents and relationships surrounding it.
The contract provides the text. The Context Graph provides a bigger picture.
See the full transaction behind every contract: from individual contracts to complete legal transactions.
An isolated contract rarely tells the whole story: an amendment may overwrite an original provision. A related agreement may change an obligation. A playbook may define the organization's preferred position. Previous negotiations may provide important commercial context.
Luminance analyzes relationships between documents and connects them into Context Graphs representing the wider legal transaction. Instead of reasoning over an isolated file, the platform can retrieve the documents and relationships surrounding it.
The contract provides the text. The Context Graph provides a bigger picture.
Give AI the context it needs to reason.
Pre-computing legal knowledge means Luminance doesn't need to rediscover every clause, entity, or legal concept from raw text before it can begin answering a question. Common legal knowledge has already been extracted and indexed.
That enables:
Faster retrieval: Known legal information can be queried directly rather than generated from scratch.
More consistent analysis: Agents and users can interrogate the same structured legal data through a common query framework.
Better context: Context Graphs allow answers to account for related documents and legal transactions rather than treating every contract as an isolated file.
Retrieval built for scale: Different retrieval techniques can be applied depending on whether the task requires a single passage, multiple documents, or portfolio-wide analysis.
The goal: give AI the right legal context before it starts reasoning.
Give AI the context it needs to reason.
Pre-computing legal knowledge means Luminance doesn't need to rediscover every clause, entity, or legal concept from raw text before it can begin answering a question. Common legal knowledge has already been extracted and indexed.
That enables:
Faster retrieval: Known legal information can be queried directly rather than generated from scratch.
More consistent analysis: Agents and users can interrogate the same structured legal data through a common query framework.
Better context: Context Graphs allow answers to account for related documents and legal transactions rather than treating every contract as an isolated file.
Retrieval built for scale: Different retrieval techniques can be applied depending on whether the task requires a single passage, multiple documents, or portfolio-wide analysis.
The goal: give AI the right legal context before it starts reasoning.
Give AI the context it needs to reason.
Pre-computing legal knowledge means Luminance doesn't need to rediscover every clause, entity, or legal concept from raw text before it can begin answering a question. Common legal knowledge has already been extracted and indexed.
That enables:
Faster retrieval: Known legal information can be queried directly rather than generated from scratch.
More consistent analysis: Agents and users can interrogate the same structured legal data through a common query framework.
Better context: Context Graphs allow answers to account for related documents and legal transactions rather than treating every contract as an isolated file.
Retrieval built for scale: Different retrieval techniques can be applied depending on whether the task requires a single passage, multiple documents, or portfolio-wide analysis.
The goal: give AI the right legal context before it starts reasoning.
Explore the architecture
From the right context to the right answer.
Model-Opinionated
A multi-model architecture routes each task to the intelligence best suited to its accuracy, speed, reliability and cost requirements, without dependence on a single model or provider.
Differential Validation
Independent judges, grounded critics and task-specific checks scrutinise high-impact outputs against evidence, improving accuracy and making uncertainty visible when human review is required.
Luna Crescent
Luminance’s proprietary LLM is purpose-built for specialist legal tasks, improving the speed and economics of high-volume contract extraction, classification and analysis.
Explore the architecture
From the right context to the right answer.
Model-Opinionated
A multi-model architecture routes each task to the intelligence best suited to its accuracy, speed, reliability and cost requirements, without dependence on a single model or provider.
Differential Validation
Independent judges, grounded critics and task-specific checks scrutinise high-impact outputs against evidence, improving accuracy and making uncertainty visible when human review is required.
Luna Crescent
Luminance’s proprietary LLM is purpose-built for specialist legal tasks, improving the speed and economics of high-volume contract extraction, classification and analysis.
Explore the architecture
From the right context to the right answer.
Model-Opinionated
A multi-model architecture routes each task to the intelligence best suited to its accuracy, speed, reliability and cost requirements, without dependence on a single model or provider.
Differential Validation
Independent judges, grounded critics and task-specific checks scrutinise high-impact outputs against evidence, improving accuracy and making uncertainty visible when human review is required.
Luna Crescent
Luminance’s proprietary LLM is purpose-built for specialist legal tasks, improving the speed and economics of high-volume contract extraction, classification and analysis.
GET STARTED
Get more from your contracts
See how Luminance can help your teams negotiate smarter, surface what matters, and keep contracts moving across the enterprise.
GET STARTED
Get more from your contracts
See how Luminance can help your teams negotiate smarter, surface what matters, and keep contracts moving across the enterprise.
GET STARTED
Get more from your contracts
See how Luminance can help your teams negotiate smarter, surface what matters, and keep contracts moving across the enterprise.
© 2026 Luminance. All rights reserved.