Dakera Playground · live API, no account

Run agent memoryin your browser

Store a memory, recall it by meaning, watch it fade, share it between agents. Every call here goes to the real Dakera API, in Python, JavaScript, Rust, Go or cURL.

Explore the playground Run it locally

9live modes
5languages
~1 msp95 vector search

What makes Dakera different

01
Your data stays on your servers
Self-hosted, with embeddings computed on your machine by ONNX Runtime. No memory, prompt or vector is sent to a third-party API.
02
One binary instead of four services
Vector store, embeddings, knowledge graph and sessions in one Rust process. No Redis, Pinecone or graph database to run beside it.
03
Memory that manages itself
Every memory carries an importance score. What agents keep using stays on top; what they stop using fades. No TTL scripts or cleanup jobs.
04
Fits the agents you already run
SDKs for Python, TypeScript, Go and Rust, an MCP server for Claude, Cursor and Windsurf, and LangChain and LlamaIndex integrations.

Live sandbox — Run makes the calls shown below against the live Dakera API. No account, no setup. Copy the code to use it in your own project.

playground.pyread-only

30 req/min · 50 mem/session · ⌘⏎ to run

importance 0.80

top_k

output.json idle

// Press ▶ Run to execute…

Live visualization

// Run a scenario to see the visualization

Compare recall strategies

Same query, three depth levels. See how top_k shapes what your agent retrieves — breadth vs. precision.

Try a query preset:

Broad Recall

top_k=10 · all results ranked

Returns up to 10 memories ranked by smart score. Best for discovery — you see everything the agent knows about this topic, including loosely related facts. Use when you want maximum context.

discovery mode

▶ Click Compare to run

Balanced

top_k=5 · top half of results

Returns the top 5 most relevant memories. Filters out peripheral context while keeping all meaningful signals. The default sweet spot for most agent use cases — not too noisy, not too narrow.

default mode

▶ Click Compare to run

Precise

top_k=3 · highest relevance only

Returns only the 3 highest-scoring memories. Maximum signal-to-noise ratio. Use when your prompt window is tight or you need the agent to act on only its strongest beliefs — ideal for critical decisions.

precision mode

▶ Click Compare to run

Smart Score = vector similarity × importance weight × temporal recency factor — a single 0–100% relevance signal combining semantic match, how critical the memory is, and how recently it was stored or accessed.

Free-form API explorer

Call any Dakera endpoint directly — craft your own request, see the raw response.

RequestPOST

Request Body (JSON)

Responseidle

// Response will appear here…

AI with memory, and without

Ask the same question to an LLM with and without Dakera memory. See exactly how persistent context changes the answer — and which memories made the difference.

Try a preset scenario:

1Seed memories

2Recall context

3LLM inference

4Compare

Without memory

LLM has no context about your data

Select a preset or type a question, then click Compare…

Generic response — no domain knowledge

With Dakera memory

LLM receives recalled memories as context

Select a preset or type a question, then click Compare…

Precise, contextual response — powered by recalled memories

Memories injected as context

AI chat with memory

Chat with an AI that actually remembers you. Every message is stored via Dakera — the AI recalls prior context to give personalized, contextual responses.

Session not started

Show raw JSON

Raw API Responses

Hybrid search tuner

Adjust the balance between BM25 keyword search and vector semantic search. See how results change in real-time.

BM25 (keyword) Vector (semantic)

vector_weight: 0.50

Seed memories, then search to see results

Show raw JSON

Entity extraction

Paste or type text. Dakera extracts named entities (people, orgs, locations, dates) with highlighted spans.

Type text above and click Extract

Show raw JSON

Multi-agent memory sharing

Three specialized agents share a memory namespace. Onboarding stores user context, Support recalls it for personalized help, Analytics surfaces insights. Zero re-introduction — instant cross-agent context via Dakera.

✓ Agent A stored context

▸

2 Agent B recalls

▸

3 Agent C analyzes

🤖

Onboarding Bot

Captures user preferences & context

Importance: 0.85

💬

Support Bot

Recalls shared context & generates response

📊

Analytics Agent

Recalls all context & surfaces insights

Show raw JSON

Knowledge graph explorer

Seed org memories, extract entities, then visualize the knowledge graph with force-directed layout. Click nodes to see details, drag to rearrange, double-click to expand, Shift+click two nodes to find shortest path. Each mode uses its own isolated agent namespace so graph data stays separate from chat, hybrid search, and other modes.

Seed memories and build graph to explore

Node Details

Show raw JSON

Now run it on your machine

Self-hosted, open core, one binary. From zero to recalled memories in under five minutes.

Get started free Read the docs Compare alternatives Deploy

Run agent memoryin your browser

Compare recall strategies

Free-form API explorer

AI with memory, and without

AI chat with memory

Hybrid search tuner

Entity extraction

Multi-agent memory sharing

Knowledge graph explorer

Now run it on your machine

Lock in founder pricing

Lock in founder pricing