Read and structure
OCR, layout, conversion, splitting, schema extraction, entities, PII, classification, summaries, translation.
Private Document Intelligence · Business Preview
One install. The whole platform.
LM-Kit One is LM-Kit as a product you deploy. Documents become structured data, corpora become cited answers, agents do the work with governed tools, and models are managed and fine-tuned in place. Nothing to assemble, nothing to send out.
Search and ground
Collections, lexical and semantic retrieval, reranking, answers carrying document and page citations.
Do the work
Named agents with pinned skills, governed tools, memory across conversations and a tool-call budget.
Manage and adapt
Import, quantize, load and switch models, and fine-tune them as tracked jobs with downloadable artifacts.
The scope, stated plainly.
This is not an inference server with a document endpoint bolted on. It is the whole LM-Kit engine, operated.
Read
Getting the content out
- OCR for scans and photographs
- Layout and reading-order recovery
- Markdown conversion, splitting, thumbnails
- Audio transcription, video frames
Understand
Turning it into fields
- Schema-constrained extraction with confidence
- Entities, personal data, categories
- Sentiment, keywords, language
- Summarize, translate, correct, rewrite
Search
A real retrieval engine
- BM25 full text with language analyzers
- Vector and hybrid with rank fusion
- Filters, facets, sorting, pagination
- Reranking, grounded answers with citations
Act
Changing the artifact
- Reviewed, permanent redaction
- PDF edit, merge, split, unlock, forms
- Archival PDF/A, image to PDF, rendering
- Agents driving any of it as tools
Adapt
Owning the model
- Catalog pulls, import, quantization
- Load and switch, hardware fit shown first
- Fine-tuning as submit, poll, cancel jobs
- Artifacts you download and keep
Operate
Running it in production
- Admin console, request history, telemetry
- Access tokens, skills, collections
- Windows service or desktop session
- Linux, macOS, air-gapped supported
Define the agent once. Every client adopts it by name.
Most stacks make every application rebuild the same prompt, the same tool wiring and the same memory handling. Here an agent is a server-side definition, and a request adopts it with one field. Change the agent and every caller follows, with no redeploy on their side.
Definition
What an agent carries
A system prompt, a pinned skill, the tools it may call, whether it remembers across conversations and where that memory lives, the model it answers with, its thinking mode, an opening greeting, and a ceiling on tool calls so a loop cannot run away.
Adoption
One field on a chat request
Name the agent and the bundle fills in wherever the request stays silent, while anything the request states explicitly wins. Clients discover what is available from the agents endpoint rather than being told out of band.
Skills
Reusable behaviour, installable
Ships with skills for summarizing, extracting data, proofreading, translating, writing email, taking meeting minutes, reviewing code and explaining simply. Author your own, install from a URL, and edit them from the console.
Tools
Executed by the server, under policy
Tools run on the server, not in the client, so what an agent can touch is an operator decision. Web search across several providers, page reading, HTTP reads, file ingestion and listing, the document operations, calculation, unit conversion and date maths. Outbound access is governed by an allowlist rather than assumed.
Fields your systems can consume, not prose.
Define the schema. The server reads the file, including scans and photographs, and returns JSON that matches it, with a confidence figure per field and null where the source does not say. The result is shaped for the next system, not for a reader.
In
Whatever arrives
PDFs, Office files, email, images and scans. OCR and layout recovery for documents that were never digital. Batched scans are classified and split into their constituent documents before anything else runs.
Out
Typed and confidence-scored
Grammar-constrained decoding means the response parses every time, so a malformed answer cannot reach your pipeline. Per-field confidence lets you route the uncertain cases to a person instead of accepting them silently.
Also
The analysis around it
Named entities, personal data detection, categorization, sentiment, keywords and language identification, plus summarization, translation, correction and rewriting when the task is about the text rather than its fields.
Every claim arrives with its page number.
Build collections from your policies, manuals, specifications, contracts and support archives. Query them by wording, by meaning, or both together. When the corpus does not support an answer, that is what comes back.
Retrieval
Lexical, semantic, hybrid
Exact wording matters for a part number, meaning matters for a policy question. Run either or both, with reranking to sharpen the shortlist, and see which strategy produced each result.
Storage
Nothing to provision
Full-text and vector search run on a local cluster that comes up with the product. PostgreSQL with pgvector and Qdrant are supported when you would rather run infrastructure you already operate.
Evidence
Checkable, not asserted
Answers carry the document, the page, the passage and the retrieval score, so a reviewer can verify a claim against the original rather than trusting it. Collections are administered from the console.
A search cluster, without the search cluster.
Teams reach for OpenSearch or Elasticsearch because they need real retrieval: BM25 with proper analyzers, filters, facets, pagination, multi-tenancy, reindexing. LM-Kit One has those, and the vector, hybrid and reranking layers on top, without a cluster to size, secure and keep alive.
Lexical
BM25 that respects language
Tunable BM25 with per-tenant text normalization and language-aware analysis, including Snowball stemming and bigram handling for CJK. Exact wording still wins when a part number or a clause reference is what the user typed.
Hybrid
Two fusion strategies, weighted
Run the lexical and vector arms together and fuse them by reciprocal rank, which ignores incomparable score magnitudes, or by a convex combination of normalized scores when you trust them. Per-arm weights are yours to set.
Shaping
The knobs a real system needs
Filters and a JSON filter grammar, facet counts, sort, offset paging, minimum score, per-document collapse, MMR diversity, recency decay with a half-life you choose, context expansion to the chunk or the page, and cross-encoder reranking blended into the final order.
Query
Understanding before retrieval
Rewrite a follow-up question against the conversation so pronouns resolve, and fan a query into several variants to widen recall before fusion. Both are switches on the request, not a separate service to run.
Tenancy
Clusters, tenants, collections
Multi-tenant by construction, with tenant promotion and demotion, per-collection configuration, queries that span several collections, more-like-this lookups, and reindexing as a tracked operation rather than a maintenance window.
Evidence
Measured, not asserted
Build golden sets, generate candidates for them, and score retrieval and answers against them in tracked evaluation runs. A slow-query log and top-query statistics show what production is actually asking.
Worth being precise: this replaces what teams run a search cluster for, but it is not wire-compatible with the OpenSearch or Elasticsearch APIs. Moving over means reindexing your corpus and porting queries to this one, not repointing a client. Storage can stay local, or run on PostgreSQL with pgvector or on Qdrant.
Change the document. Change the model.
Detection is only useful if something happens next, and a general model is only useful until your vocabulary matters. Both of those are operations the platform runs.
Redact
Reviewed, then permanent
Find personal data across every page, group repeated values, and present each finding with its evidence for a person to keep or remove. Approved regions are deleted from the file rather than drawn over. Reopen the output and verify.
Operate
The unglamorous PDF work
Inspect, search, render, edit, merge, split, unlock, read and fill forms, extract attachments, convert images to PDF and documents to archival PDF/A. Transcribe audio and pull frames from video while you are there.
Train
Fine-tuning as a managed job
Submit a training run, poll it, cancel it, collect the artifact. Adapting a model to your vocabulary becomes an operation the platform tracks rather than a script someone runs on a spare workstation. Training data never leaves, which is the one thing a hosted provider cannot offer.
Four interfaces, one private engine.
The work happens on your side of the boundary. Calling it should not be a migration project.
Native
The full REST surface
Every capability above is an endpoint, including the document, extraction, search, agent, training and model-management operations that no general chat API models. This is the platform at its widest.
Compatible
Three dialects, not three shims
Deep enough that real tooling drives it unmodified, including model lifecycle rather than chat alone. See the breakdown below.
MCP
Private tools for AI assistants
A curated catalog of document and knowledge tools over the Model Context Protocol. The assistant asks for an extraction, a search or a redaction; the server opens the file, works locally, and returns only the result of the tool you allowed.
How the MCP integration worksConsole
A surface for people
A workbench for trying capabilities against real documents, a training view, collection administration, and an admin console covering models, inference shape, access tokens, skills, request history, telemetry and logs.
Your existing clients, pointed somewhere private.
Compatibility is usually a chat endpoint and a promise. These dialects go deep enough that the real command-line tools drive the server unmodified, including pulling, creating and pushing models, which is the part shims normally skip.
OpenAI
Beyond chat completions
Chat completions, completions, embeddings and models, plus files, vector stores with their file operations, and the Responses API including input items. That means OpenAI-shaped retrieval and agentic flows work against a private deployment, not only single turns.
Ollama
The whole surface, lifecycle included
Chat, generate, embed, tags, show, ps and version, and the model management verbs too: pull, push, create, copy, delete and blob upload. The stock CLI and anything written against it work as they are, and a multi-gigabyte model upload is a tested path.
Anthropic
Messages, with thinking
The Messages endpoint in the Anthropic shape, so SDKs and coding agents built on it keep working while inference stays inside your network.
In practice
Tested against real clients
Compatibility that is only unit-tested tends to break on contact with an actual tool. These paths are exercised end to end with the stock command-line clients, including an agentic run that called a shell tool and verified the file it produced.
The dialects are the on-ramp, not the ceiling. Once a client is pointed at LM-Kit One, the native REST surface adds the document, extraction, search, agent and training operations none of these APIs can express.
Closed first, opened deliberately.
A private AI product that is reachable by default is not private. LM-Kit One starts shut and expects you to make each decision to open it.
Network
Loopback until you say otherwise
A fresh install listens on the local machine only. Exposing it to a network is an explicit change, and anonymous access over a network is refused rather than warned about.
Access
Tokens hashed at rest
API tokens are stored hashed, not in plain text. Sessions and the files they can reach are bound to the authenticated owner, so one session cannot read another owner documents.
Egress
Outbound access is a policy
Tools that reach the internet are governed by an allowlist rather than assumed. Once models are present, documents, indexing, retrieval, training and inference need no outbound connection at all, and air-gapped operation is a supported mode.
Business Preview, and what that means.
LM-Kit One is in Business Preview ahead of general availability. We would rather be precise about that than let you find out later.
You get
A working platform, and the team
Everything on this page is built and running, not planned. Preview participants work directly with the engineering team, and what they hit in their own environment is what shapes the release.
Not yet
Support commitments
No service-level agreement, no response-time commitment, and interfaces can still change before general availability. If you need contractual assurance today, talk to us about timing rather than deploying on preview terms.
Next
General availability
Supported builds, long-term-support versions, security advisories and a response-time commitment arrive with the release. Preview participants carry their deployment across rather than starting again.
Free for small teams. Commercial above the line.
One license, both products, no activation key. Nothing checks a license at runtime. Evaluation and development are free at any company size, with no time limit.
Free
$0no key, no expiry
The complete SDK and the complete server, including commercial use and redistribution, for small companies. Evaluation and development stay free at any size.
- Under $1M USD annual gross revenue
- 10 or fewer employees
- No more than $3M USD raised from outside investors
- Always free: personal, education, nonprofits, open source
Professional
Customannual, scaled to scope
Required above the thresholds. Scaled to deployment size, never metered by tokens, seats or end users. Carries the assurance a production deployment needs.
- Commercial use and redistribution at any scale
- Long-term support builds and security patches
- Support with response-time commitments
- Unlimited developers and end users
Private Document Intelligence