TelloDB Core is open source

Memory that keeps up with the person.

TelloDB turns conversations into durable, inspectable memory. It filters out what is no longer true. One Rust engine for recall, temporal truth, and deterministic answers.

Start with TelloDB View quickstart

One Rust binary · self-hosted or managed · no model lock-in

User memoryentity / amina-k

synced

May 12

Lives in Karachi

Superseded after a newer location was confirmed.

retired

Jul 28

Moved to Tokyo

Extracted from conversation · confidence 0.97

active

Jul 28

Prefers jasmine tea

Preference linked to food and beverage profile.

active

Agent asks

Where does Amina live now?

Tokyoresolved from 2 facts in 4.2 ms

Live product demo

See memory update in real time.

Experience TelloDB's real-time ingestion and recall loop. Store a fact, then retrieve it across model contexts.

STEP 1

Ingestion Layer

Write Pipeline

Demo Memory Session Key

Memories are stored and queried using this unique key. It isolates your demo session and persists across reloads.

Raw Memory Input string

Ingest Latency (Engine)

---

Ingest Latency (Network RTT)

---

STEP 2

Truth Retrieval

Read Substrate

Memory Query (Semantic Search)

Query Latency (Engine)

---

Query Latency (Network RTT)

---

Retrieved Memory Hits

No memories retrieved yet. Ingest a fact first.

Why memory matters

Chat history is not memory.

Similarity search can recover a passage, but it cannot decide whether that passage is still true. TelloDB adds time, relationships, and supersession to retrieval.

Standard Vector DB

Amnesiac & Static
Retrieves conflicting data from 2 years ago exactly like data from 2 minutes ago. No concept of evolving truth.
Fails at Counting
Cannot accurately aggregate or count facts (e.g. 'How many cars do I own?'). Relies entirely on the LLM to do math.
Bloated Storage
Stores every single conversational 'uh' and 'um' instead of maintaining a clean, structured user profile.

TelloDB Memory Engine

Fact Supersession (Temporal Truth)
When life changes (e.g. moving from NYC to SF), TelloDB marks the old fact as stale, ensuring the LLM always gets the latest truth.
Deterministic Aggregation
Built-in execution layer accurately computes numeric and temporal queries before hitting the LLM, fixing benchmark failures.
Predict-Calibrate Profile
Distills thousands of words into compact, continuous user profiles. We track the deltas, you save on context windows.

Recall Precision Benchmarks

LongMemEval-S Suite

100%75%50%25%0%

68%

Standard RAG

95%+

TelloDB

TelloDB resolves standard vector search failures. In memory tasks with high fact-density, the local hybrid architecture ensures precise recall.

Temporal Fact Evolution

Supersession State

2025 (STALE)"Living in NYC"

Today (ACTIVE)"Moving to SF"

Fact Invalidation Complete

When newer truths supersede older context, the engine automatically tags prior states as stale, filtering them out from active agent context.

How memory stays useful

Keep the signal.
Retire the noise.

TelloDB turns conversations into current facts, preferences, and relationships that can be queried without replaying an entire transcript.

Time-Awareness

I used to love coffee, but now I only drink tea. TelloDB does not hallucinate your old preferences. It updates your profile in real time.

Fact Distillation

Our engine automatically discards greetings and filler, keeping only the high-value semantic facts that actually matter for personalization.

User: "Hey! I just bought a white Mercedes!"

Raw Chat

Distillation Kernel

Rust Semantic Filter v0.1.0

Fact: User owns Mercedes (White)

Committed to memory lattice

Shared user context

One memory,
every application.

Give every agent the same current understanding of the user, without locking that context inside one conversation or model.

The First Spark (May 12)

Hey! I just bought a white Mercedes! What should I do first?

GPT-4o detects: User Ownership → Vehicle: Mercedes (White)

TelloDB Ingests

Fact Integration

Fact: Owns Mercedes

Context: Initial Purchase

3 Months Later (Aug 20)

What was that maintenance tip for my car?

Claude 3.5 recalls: "For your white Mercedes, I recommend..."

Memory primitives

Built for production agents.

TelloDB replaces complex, slow orchestration chains with a unified, high-performance memory engine.

Rust Performance Core

Engineered in Rust with sub-100ms p99 query latencies, zero GC pauses, and compiled as a single air-gapped binary.

Fact Supersession

Time-aware rankings and TTL decay policies automatically tag older context as stale when new conflicting facts arrive.

Deterministic Analytics

Built-in arithmetic and count aggregation calculated directly on the database B-Tree indexes before LLM delivery.

Multi-Model Continuity

Switch cognitive backends (GPT-4, Claude 3.5, Llama 3) without losing memory state or query syntax history.

OpenAI Proxy Path (Coming Soon)

Drop-in proxy gateway that automatically injects relevant context into OpenAI-compatible system instructions. In development.

Graph Knowledge Base

Self-organizing RDF typed graph representations mapping relationships between user sessions and profile history.

Recall Engine Latency

<100ms

Average p99 Recall

Built natively in Rust. Delivers sub-100ms queries under heavy semantic and full-text loads.

Runtime Configuration

engine: TelloDB
routes: /ingest /query/semantic /query/temporal /memory
indexes: hnsw + bm25 + graph lineage
policy: ttl + decay + supersession
sdk: python + javascript

ProductionLocal-FirstModel-Agnostic

Memory pipeline

From conversation to usable context.

TelloDB turns raw conversation into concise, current context that agents can retrieve and use.

IngestRaw Events

Distill & StoreMemory controller

HNSWBM25GRAPH

Final TruthGrounded Context

Intent-Aware Filtering

Automatically detects if the user is asking for numbers, preferences, or narrative history.

Neural Reranking

Applies a secondary precision pass to ensure the top-k candidates are semantically perfect.

Deterministic Compute

Computes aggregates (sums, counts) before delivery, preventing LLM arithmetic errors.

Relationship explorer

Inspect context as a graph.

Explore how TelloDB connects entities, preferences, and facts. Drag nodes to inspect relationships and see how newer information supersedes stale context.

Active Truth

Superseded Memory

Nodes represent discrete semantic facts, preferences, and entities stored within the Rust engine.

Red nodes indicate **superseded memories**—stale data that has been automatically invalidated by more recent truths.

Adoption path

Start local.
Scale when ready.

The product has a clear progression: ingest fidelity, retrieval intelligence, and operational reliability.

Phase 01

Ingest and Distill

Raw events are normalized, deduplicated, and expanded into durable memories with lineage.

Companion memories
Dedup table
Graph relationships

Phase 02

Retrieve and Rerank

Semantic and lexical candidates are fused, reranked, then filtered by temporal policy before response.

HNSW + BM25
Cross-rerank
RRF + policy filters

Phase 03

Ship and Operate

Teams deploy one memory engine surface from local bench runs to hosted multi-tenant workloads.

SDK parity
Benchmarked quality
Operational playbooks

Book A Session

Schedule a 30-minute walkthrough.

Discuss database architecture, memory integration, and private deployment setups for your agentic applications.

Open Calendly Email Instead

Give every agent
durable memory.

One binary, five memory substrates, zero lock-in. Start building persistent, self-improving agents today.

Get Started Free Book a Demo

Deployment options

Engine and Platform

TelloDB Core

Open Source Engine

The Rust-powered temporal memory engine that runs anywhere. Hybrid vector + BM25 search, knowledge graph traversal, deterministic analytics, and fact supersession — all in a single binary. No vendor lock-in.

▸Single binary — download and run locally
▸4 HNSW indexes + BM25F + redb KV + typed graph
▸Candle / ONNX embeddings, CPU or GPU targets
▸Embed directly in your custom agent architecture

Read the Docs Contact Us

TelloDB Platform

Managed Cloud Service

A full SaaS experience on top of the core engine. Deploy clusters in one click, manage your team, track usage with analytics, explore knowledge graphs visually, and never worry about infrastructure.

▸One-click cluster provisioning — no config files
▸Stripe integration — pay-as-you-go or flat-rates
▸Team management with invites and RBAC roles
▸Graph explorer, telemetry, and visual playground

Get Started Free Explore Platform Documentation

Works with your stack

One memory layer, every model.

We provide the tooling to make persistent memory a first-class citizen in your development workflow, from local testing to global scale.