Quick Start

Get up and running with Gigamemory in under 5 minutes. All you need is an API key and an HTTP client.

Authentication

Every request to the Gigamemory API requires an API key sent via the X-Memory-Key header. Generate your key from the dashboard.

X-Memory-Key: gm_key_xxxxxxxxxxxxxxxx

Ingest a Memory

Send a piece of information — an "episode" — to your memory store. Gigamemory automatically extracts entities and relationships, building a graph that connects every fact you add.

Endpoint

POST /v1/episodes

Request Body

Example

curl -X POST https://api.gigamemory.ai/v1/episodes \
  -H "X-Memory-Key: gm_key_xxxxxxxxxxxxxxxx" \
  -H "Content-Type: application/json" \
  -d '{
    "text": "Acme Corp was founded in 2020 by Jane Smith. Jane is the CEO and has a background in enterprise software.",
    "metadata": {
      "source": "company_profile",
      "tags": ["acquisition", "leadership"]
    }
  }'

import requests

response = requests.post(
    "https://api.gigamemory.ai/v1/episodes",
    headers={
        "X-Memory-Key": "gm_key_xxxxxxxxxxxxxxxx",
        "Content-Type": "application/json"
    },
    json={
        "text": "Acme Corp was founded in 2020 by Jane Smith. "
                "Jane is the CEO and has a background in enterprise software.",
        "metadata": {
            "source": "company_profile",
            "tags": ["acquisition", "leadership"]
        }
    }
)

print(response.json())

const response = await fetch("https://api.gigamemory.ai/v1/episodes", {
  method: "POST",
  headers: {
    "X-Memory-Key": "gm_key_xxxxxxxxxxxxxxxx",
    "Content-Type": "application/json"
  },
  body: JSON.stringify({
    text: "Acme Corp was founded in 2020 by Jane Smith. "
        + "Jane is the CEO and has a background in enterprise software.",
    metadata: {
      source: "company_profile",
      tags: ["acquisition", "leadership"]
    }
  })
});

const result = await response.json();
console.log(result);

Response

{
  "id": "ep_abc123def",
  "entities_extracted": 2,
  "relationships_extracted": 1,
  "created_at": "2026-06-13T10:30:00Z"
}

The response tells you how many entities (people, companies) and relationships (founded by, is CEO of) were extracted and stored in the graph.

Search Memories

Query your memory graph using natural language. Results include both relevant facts and the graph relationships that connect them.

Endpoint

POST /v1/search

Request Body

Example

curl -X POST https://api.gigamemory.ai/v1/search \
  -H "X-Memory-Key: gm_key_xxxxxxxxxxxxxxxx" \
  -H "Content-Type: application/json" \
  -d '{
    "query": "Tell me about Acme Corp and its leadership",
    "limit": 5
  }'

import requests

response = requests.post(
    "https://api.gigamemory.ai/v1/search",
    headers={
        "X-Memory-Key": "gm_key_xxxxxxxxxxxxxxxx",
        "Content-Type": "application/json"
    },
    json={
        "query": "Tell me about Acme Corp and its leadership",
        "limit": 5
    }
)

results = response.json()
for r in results["results"]:
    print(f"{r['entity']}: {r['fact']}")

const response = await fetch("https://api.gigamemory.ai/v1/search", {
  method: "POST",
  headers: {
    "X-Memory-Key": "gm_key_xxxxxxxxxxxxxxxx",
    "Content-Type": "application/json"
  },
  body: JSON.stringify({
    query: "Tell me about Acme Corp and its leadership",
    limit: 5
  })
});

const results = await response.json();
console.log(results);

Response

{
  "results": [
    {
      "entity": "Acme Corp",
      "fact": "Acme Corp was founded in 2020 by Jane Smith",
      "relevance": 0.94
    },
    {
      "entity": "Jane Smith",
      "fact": "Jane Smith is CEO of Acme Corp",
      "relevance": 0.91
    }
  ],
  "graph": {
    "relationships": [
      { "source": "Jane Smith", "relation": "founded", "target": "Acme Corp" },
      { "source": "Jane Smith", "relation": "is CEO of", "target": "Acme Corp" }
    ]
  }
}

Unlike vector-only systems that return similarity matches, Gigamemory returns the graph relationships connecting your results. You get why something is relevant, not just that it is.

Check Usage

Monitor your current entity count and API usage for the billing period.

Endpoint

GET /v1/usage

Example

curl https://api.gigamemory.ai/v1/usage \
  -H "X-Memory-Key: gm_key_xxxxxxxxxxxxxxxx"

import requests

response = requests.get(
    "https://api.gigamemory.ai/v1/usage",
    headers={"X-Memory-Key": "gm_key_xxxxxxxxxxxxxxxx"}
)

print(response.json())

const response = await fetch("https://api.gigamemory.ai/v1/usage", {
  headers: {
    "X-Memory-Key": "gm_key_xxxxxxxxxxxxxxxx"
  }
});

const usage = await response.json();
console.log(usage);

Response

{
  "total_entities": 1247,
  "entity_limit": 10000,
  "api_calls_this_period": 3421,
  "period_reset": "2026-07-01T00:00:00Z"
}

Rate Limits

Rate limits vary by plan and are enforced on a per-API-key basis. Every response includes rate limit headers so you can track your usage programmatically.

See the Pricing FAQ for plan details and upgrade information. If you hit a rate limit, the API returns HTTP 429 with a Retry-After header indicating when to retry.

Entity Caps

Each plan has a maximum number of entities you can store. An entity is a distinct object or concept in your memory graph — a person, company, product, document, or any other discrete item your agent learns about.

Relationships between entities do not count toward the entity cap. You can have an unlimited number of relationships connecting your stored entities.

Pricing FAQ

What counts as a write?

A write is any API call that creates or updates a memory entity or relationship. Reads (queries and retrievals) are unlimited at every tier.

What counts as an entity?

An entity is a distinct object or concept stored in your memory graph — a person, company, product, document, or any other discrete item that your agent learns about.

Are there overage charges?

No. If you exceed your entity limit, you'll be prompted to upgrade to the next tier. No surprise bills, no per-request overage fees.

Is graph memory really included at every tier?

Yes. Unlike competitors who charge 10-20x more for graph capabilities, Gigamemory includes graph memory at every tier from Starter to Enterprise.

Can I try before I buy?

Yes. Sign up at the dashboard for a free trial with full access to all features.

What about data residency?

Enterprise customers can request co-location in specific regions to keep data close to their agents. Contact [email protected] for details.