The universal memory API for the AI era
Context engine for your app. Personalise LLMs for your
users.
Built for developers who ship.
Context
is everything
Without it, even the smartest AI is just an expensive chatbot
- $ init multimodal support
- Websites: JS & rate limits are messy
- PDFs: OCR fails, extraction inconsistent
- Authentication tokens expire constantly
-
- $ init vector_database
- Way too expensive. Time to switch.
- Painfully slow. Let's try another.
- Won't scale. Back to square one.
- Maintenance nightmare. Need alternatives.
-
- $ choose embedding_model
- Which model fits your use case?
- Confusing performance tradeoffs
- Can't keep up with new releases
-
- $ handle format_parsing
- Markdown: Tables break everything
- HTML: Scripts and styles interfere
- PDF: Layout ruins extraction
-
- $ calculate scaling_costs
- Costs explode at production scale
- Performance degrades as data grows
- Engineering hours pile up fast
-
- $ setup connection_sync
- Sync failures between data sources
- API rate limits during large syncs
- Images: Need vision models now?
- Audio/Video: Transcription costs soar
-
- $ init multimodal support
- Websites: JS & rate limits are messy
- PDFs: OCR fails, extraction inconsistent
- Authentication tokens expire constantly
-
- $ init vector_database
- Way too expensive. Time to switch.
- Painfully slow. Let's try another.
- Won't scale. Back to square one.
- Maintenance nightmare. Need alternatives.
-
- $ choose embedding_model
- Which model fits your use case?
- Confusing performance tradeoffs
- Can't keep up with new releases
-
- $ handle format_parsing
- Markdown: Tables break everything
- HTML: Scripts and styles interfere
- PDF: Layout ruins extraction
-
- $ calculate scaling_costs
- Costs explode at production scale
- Performance degrades as data grows
- Engineering hours pile up fast
-
- $ setup connection_sync
- Sync failures between data sources
- API rate limits during large syncs
- Images: Need vision models now?
- Audio/Video: Transcription costs soar
-
We've seen what it's like to build memory infrastructure the hard way — so we built supermemory to make it disappear.
Edit one line.
Get longer threads,
cost savings
memory.
Just add api.supermemory.ai/v3 to your OpenAI base URL — and get automatic long-term context across conversations.
import OpenAI from "openai"
const client = new OpenAI({
baseUrl: "https://api.supermemory.ai/v3/https://api.openai.com/v1/"
})Unlock the Full Potential of Your Data
const response = await fetch('https://api.supermemory.ai/v3/memories', {
method: 'POST',
headers: {
'Authorization': 'Bearer sm_ywdhjSbiDLkLIjjVotSegR_rsq3ZZKNRJmVr12p4ItTcf'
},
body: JSON.stringify({
content: 'You can add text',
// or a url https://example.com
// or pdfs, videos, images. https://example.com/page.pdf
metadata: {
user_id: '123'
}
}),
})
const data = await response.json();const response = await fetch('https://api.supermemory.ai/v3/memories', {
method: 'GET',
headers: {
'Authorization': 'Bearer sm_ywdhjSbiDLkLIjjVotSegR_rsq3ZZKNRJmVr12p4ItTcf',
},
body: JSON.stringify({
q: "What's my name?"
})
})
const data = await response.json()
const response = await fetch('https://api.supermemory.ai/v3/connections/onedrive', {
method: 'POST',
headers: {
'Authorization': 'Bearer sm_ywdhjSbiDLkLIjjVotSegR_rsq3ZZKNRJmVr12p4ItTcf',
}
});
const data = await response.json();
Build the memory layer your product deserves
Enterprise-Grade Performance at Any Scale
supermemory is built to handle billions of data points with low-latency retrieval — whether you're indexing documents, video, or structured product data.
Seamless Integration Across Teams & Tools
Connect directly to your existing stack — from Notion to Google Drive to custom CRMs — with flexible APIs and SDKs that let every team tap into memory instantly.
Secure by Design.
Fully Controllable.
Deploy supermemory in the cloud, on-prem, or directly on-device — with full control over where and how your data is stored.
It just clicks with your stack
Model-interoperable APIs
supermemory works with any LLM provider. So you can keep switching, without lock-in. Switch between models. keep your memory.
Sub-400ms latency at scale
supermemory is built for speed and scale. We re-imagined RAG to be faster and more efficient.
Best in class performance
supermemory delivers stronger precision and recall at every benchmark. And it's ridiculously easy to start.
supermemory vs major memory provider
Works with AI SDK, Langchain, and more
supermemory works with any LLM provider. So you can keep switching, without lock-in.
Add context to your agentic apps with few lines of code
supermemory provides SDKs to make integration as simple as possible
Trusted by Open Source, enterprise, and more than
of you
