The managed retrieval layer for enterprise AI.
VectorForgeAI turns your organisation's documents into semantic search, grounded conversational AI, and an MCP endpoint, delivered as one fully managed API. There is no vector database or embedding pipeline for your team to operate.
14-day trial on every plan. No credit card required.
Deployment profile
Fully managed- Data residency
- EU data isolation
- Encryption
- TLS 1.3 in transit, AES-256 at rest
- Hosting
- ISO/IEC 27001 certified data centres
- Retention
- Conversations deleted after 30 days
- Interfaces
- REST API and MCP server
- Identity
- SAML and Okta on Enterprise
-
refunds.md chunk 2 0.89
Enterprise customers may request a refund within 30 days of initial subscription.
-
billing-faq.md chunk 0 0.71
Refund requests are submitted through the account management portal.
-
terms.md chunk 7 0.42
Subscription fees are billed monthly in advance and are non-transferable.
-
onboarding.md chunk 1 0.19
Invite teammates from the team settings screen.
Results scoring below 0.35 are filtered out before they reach your application.
User
How do refunds work for enterprise plans?
Assistant, grounded in 2 sources
Enterprise customers can request a refund within 30 days of their initial subscription. Requests are submitted through the account management portal.
Sources
- TLS 1.3 Encryption in transit
- AES-256 Encryption at rest
- ISO/IEC 27001 Certified data centres
- GDPR ready EU data isolation
- SAML and Okta Single sign-on on Enterprise
Platform
One platform for search, conversation, and model access
Every capability reads from the same document collections, so content is ingested once and governed in one place, whichever interface your teams and assistants use.
Semantic search
Query collections by meaning. Results are reranked, filtered, and returned as documents with similarity scores, never raw vectors.
Search referenceConversation API
Stateful chat that keeps the thread, retrieves from your collections, calls your functions, and scores each outcome.
Explore conversationsLLM completions
Generate text with your own context through the Responses API, on a base or Pro model tier.
Completions referenceMCP server
Make collections available to Claude, ChatGPT, and other MCP clients using the tokens you already have.
MCP guideManaged ingestion
From raw text to governed retrieval in four stages
Every document passes through the same pipeline: the one your engineering team would otherwise have to build, operate, and maintain.
-
1
Ingest
POST a body, an identifier, and any metadata you want to filter on later. Re-sending an identifier replaces the document in place.
-
2
Enrich and chunk
Text is converted to Markdown, summarised into problem and solution pairs, then split into 100–200 word chunks with configurable overlap.
-
3
Embed
Each chunk and pair is embedded with a managed embedding model. Identical text is embedded once and reused wherever it appears.
-
4
Retrieve
Queries are embedded, matched by similarity, reranked, and filtered. Your application receives documents and scores, not vectors.
For engineering teams
Three API calls cover the platform
Store, search, and converse through a consistent REST interface. Integration work stays in your application, not in infrastructure.
- Every request authenticates with a bearer token and a team token.
- Requests and responses are JSON, described by a published OpenAPI schema.
- Documents are keyed by your own identifier, so updates replace in place.
- Conversations persist server-side; you store only the conversation id.
curl -X POST https://api.vectorforgeai.com/v1/collections/abc123/documents \
-H "Authorization: Bearer $API_KEY" \
-H "Team-Token: $TEAM_TOKEN" \
-H "Content-Type: application/json" \
-d '{
"identifier": "refunds",
"title": "Refund policy",
"body": "Enterprise customers may request a refund within...",
"metadata": { "department": "Finance" }
}'curl "https://api.vectorforgeai.com/v1/collections/abc123/documents/search?query=refunds&limit=3" \
-H "Authorization: Bearer $API_KEY" \
-H "Team-Token: $TEAM_TOKEN"
# Each document is returned with the similarity score
# that matched it, highest first.curl -X POST https://api.vectorforgeai.com/v1/messages \
-H "Authorization: Bearer $API_KEY" \
-H "Team-Token: $TEAM_TOKEN" \
-H "Content-Type: application/json" \
-d '{
"user": "user_12345",
"message": "How do refunds work for enterprise plans?",
"collections": ["abc123"]
}'
# Returns a conversation_id. Send it back to continue
# the thread with its history intact.Model Context Protocol
Give Claude and ChatGPT governed access to your knowledge
The collections you already search through the API are also available as an MCP server. Assistants read your approved documentation directly, instead of employees pasting it into prompts.
- No extra setup
- Works with the collections you already have.
- Predictable metering
- One search credit per search_docs call.
- Existing credentials
- Authenticates with your API and team tokens.
$ claude mcp add --transport http "VectorForgeAI" \
"https://api.vectorforgeai.com/mcp" \
--header "Authorization: Bearer $API_TOKEN" \
--header "Team-Token: $TEAM_TOKEN"Solutions
Where organisations put it to work
Customer support
Support assistants
Answer customer questions from your help centre, with the source document attached to every answer. Ratings flow back through the feedback endpoint, so you can see what is working.
Internal knowledge
Search over internal documentation
Handbooks, runbooks, and specifications become searchable by meaning. Colleagues find the paragraph they need without knowing the file name.
Platform teams
Retrieval for your own models
Use search on its own and pass the chunks to whichever model you already run. The retrieval layer is independent of what generates the answer.
Security and governance
Built to pass your security review
Clear answers to the questions a security questionnaire asks first: where data lives, how it is protected, and how long it is kept.
- Data residency
- Customer data is isolated in the EU, with processing designed around GDPR.
- In transit
- TLS 1.3, falling back to TLS 1.2 for older clients.
- At rest
- AES-256 encryption in ISO/IEC 27001 certified data centres.
- Retention
- Conversations are deleted automatically after 30 days. Closing an account removes everything.
- Search terms
- Stored as SHA-256 hashes, so query text is not kept alongside the vectors.
- Enterprise assurance
- SAML and Okta single sign-on, an SLA, and access to the SOC 2 report.
Pricing
Plans that scale with your deployment
Every plan starts with a 14-day free trial. Exceeding an allowance never interrupts your application: additional usage is billed at published rates, and you can set a cap.
Basic
For small teams starting out
$69 per month
Start 14-day free trial- 300 documents
- 2,000 chat messages
- 50,000 search queries per month
- Community support
Pro
For growing businesses
$399 per month
Start 14-day free trial- 800 documents
- 10,000 chat messages
- 500,000 search queries per month
- Priority email support and SLA
Enterprise
For large organisations
Custom
Contact sales- Unlimited documents, messages, and search
- Dedicated account manager
- SAML and Okta single sign-on
- SOC 2 report and SLA
Compare plans
| Feature | Basic | Pro | Enterprise |
|---|---|---|---|
| Usage | |||
| Documents | 300 | 800 | Unlimited |
| Chat messages | 2,000 | 10,000 | Unlimited |
| Search queries per month | 50,000 | 500,000 | Unlimited |
| LLM tokens in | 1,000,000 | 5,000,000 | Unlimited |
| LLM tokens out | 500,000 | 2,500,000 | Unlimited |
| Analytics | |||
| Analytics dashboard | Standard | Advanced | Advanced |
| Support and assurance | |||
| Support | Community | Priority email | Dedicated account manager |
| SLA | — | ||
| SAML and Okta | — | — | |
| SOC 2 report | — | — | |
Usage beyond your allowance
Overage is billed per unit at the rates shown. Set a budget per usage type in the dashboard to cap spend.
- Documents
- $2.00
- per 100
- Search
- $0.50
- per 1,000 queries
- Messages
- $35.00
- per 1,000
- Tokens in
- $2.00
- per 1M, or $8.00 on Pro models
- Tokens out
- $6.50
- per 1M, or $26.00 on Pro models
What is VectorForgeAI?
A managed API that turns your text into a searchable, conversation-ready knowledge base. You send documents and receive semantic search, conversations grounded in those documents, and an MCP endpoint. Storage, embedding, and retrieval are operated by us.
What content can we send to the API?
Text. Each document is a JSON request with an identifier, an optional title, the body as a string, and any metadata you need. There is no file upload endpoint yet, so text should be extracted from PDFs or Word files before the call.
How is our content processed?
Each document is converted to Markdown, summarised into problem and solution pairs, then split into chunks of roughly 100 to 200 words. Overlap between chunks is configurable per collection and defaults to 17%. Every chunk is embedded and indexed.
Do we need to select an embedding model?
No. Embedding is fully managed, and the model is chosen for the balance of retrieval quality and cost. You never call it directly and it is never billed separately.
Which models generate answers?
The Conversation API runs on a current frontier model and moves to a smaller one once a thread grows past roughly 75,000 tokens, so long conversations remain cost-effective. The Responses API lets you choose a base or a Pro tier, priced differently. Models are kept current as better ones become available.
What happens when we exceed a plan limit?
Your application keeps working. Usage above your allowance is billed at the published overage rates, and you can set a budget per usage type in the dashboard to cap spend.
Can we evaluate it before purchasing?
Yes. Every plan starts with a 14-day free trial, and no credit card is required to begin.
How is our data protected?
Data in transit uses TLS 1.3, and data at rest is encrypted with AES-256 in ISO/IEC 27001 certified data centres. Conversations are deleted after 30 days, search terms are stored as SHA-256 hashes, and closing an account removes everything. The security overview has the full detail.
Evaluate VectorForgeAI with your own documents
Start a 14-day trial without a credit card, or speak with our team about Enterprise requirements such as single sign-on, an SLA, and the SOC 2 report.