The Living Context Engine.

Context infrastructure for AI agents. Adaptive memory, semantic graph, sub-millisecond retrieval. Ships as a single .feather file. Open source, MIT, zero-server.

View on GitHub

Open source · MITRust + C++ core<1ms retrieval

What we do

One engine, every capability.

Context infrastructure for AI agents. Adaptive, semantic, sub-millisecond. Shipped as a single file you can drop into any stack today.

Feather CoreAvailable · v0.10.0

Open source. Embedded. Yours.

Zero-server. Single .feather file. Python + Rust SDK. Ships in 5 minutes.

Embedded in-process
MIT licensed
Semantic + graph
SIMD AVX2/AVX512

Install now

Feather CloudComing Q3 2026

Managed. API-first. Global.

Your context layer, delivered globally. Keep your data in your VPC if you want to.

Managed API
Horizontal scale
VPC deployment
Usage-based pricing

The Context Stack

Five layers. Complete context.

Most memory solutions give you one layer. Feather gives you all five, working together as one engine.

[ 01 ]Adaptive Memory

Memory that ages gracefully.

Recall-based stickiness combined with time decay. Frequently accessed knowledge resists aging. Stale information fades automatically. No cron, no curation.

[ 02 ]Context Graph

Typed edges. Real reasoning.

Weighted, directional edges with BFS traversal. Your knowledge doesn't live as isolated points. It becomes a graph the engine can walk.

[ 03 ]Semantic Search

Sub-millisecond retrieval.

HNSW graph index, M=16, ef=200. Similarity kernels hand-written for AVX2 and AVX512. The .feather binary format is zero-copy. Memory-mapped, not parsed.

[ 04 ]Metadata Intelligence

Entity and attribute aware.

Namespaces, entities, attributes, importance. Filter, scope, and rerank with first-class metadata. Not bolted on, baked in.

[ 05 ]Deploy Anywhere

Embedded or managed.

Start with a single .feather file in-process. Scale to Cloud when you're ready. Same engine, same semantics, your surface of choice.

Under the hood

Knowledge that evolves, not just stores.

Not another vector database. A custom engine, built from scratch in C++ and Rust, for one job: holding the living context your agents actually need.

01 · ADAPTIVE DECAY

Memory that ages gracefully.

Every record tracks recall count, last access, and inherent importance. At query time, three scores combine into one. No cron, no eviction queue.

▸ stickiness = 1 + log(1 + recall_count)

▸ effective_age = age_days / stickiness

▸ final_score = similarity × recency × importance

02 · GRAPH ENGINE

Typed edges. Real reasoning.

Weighted, directional edges with BFS traversal. Your knowledge doesn't live as isolated points. It becomes a graph the engine can walk.

informed_bycontradictsBidirectional indexBFS via context_chainOntology-aware

03 · SIMD SEARCH

Sub-millisecond retrieval.

HNSW graph index, M=16, ef=200. Similarity kernels hand-written for AVX2 and AVX512. Zero-copy .feather binary. Memory-mapped, not parsed.

10K0.3ms

100K0.9ms

1M3.2ms

10M9.8ms

Developer Experience

Setup in 5 minutes.

One import. One file. Semantic search and graph traversal in a single call. No infra, no setup ceremony. Just open a .feather and start adding context.

pip install feather-db
Embedded. No server to deploy
Vectors + graph in one API
Works with LangChain, LangGraph, CrewAI

quickstart.py

# Install
pip install feather-db

import feather_db

# Open an embedded .feather file (or create one)
db = feather_db.DB.open("context.feather", dim=768)

# Attach rich metadata: namespace, entity, attributes, importance
meta = feather_db.Metadata()
meta.importance = 0.85
meta.set_attribute("type", "campaign_brief")

# Add a vector + its metadata
db.add(id=1001, vec=embed("your context"), meta=meta)

# Connect it into the knowledge graph
db.link(from_id=1001, to_id=1002, rel_type="informed_by", weight=0.9)

# Semantic search + 2-hop graph traversal, in one call
chain = db.context_chain(query_vec, k=5, hops=2)

Plugs into every stack you already use

Python

Rust

LangChain

LangGraph

CrewAI

OpenAI

Anthropic

Gemini

Vercel AI SDK

Deployment

Deploy your way.

Start embedded. Scale to the cloud when you're ready. The context layer is always yours. Same engine, same semantics, your choice of surface.

Aspect	Feather Core	Feather Cloud
Status	Available now	Coming Q3 2026
Deployment	In-process, single file	Managed API
Latency	<1ms	<50ms
Data	100% yours, on disk	Your VPC option
Scale	Single node	Horizontal, auto
Ops	Zero. It's a file	Fully managed
Price	Free · Open source	Usage-based

Use cases

Built for every context-hungry system.

AI Agents

Memory that updates as the agent acts.

Agents fail when their context is stale. Feather writes back every retrieval, strengthens what worked, fades what didn't, so the next turn starts smarter.

No hallucinations from outdated context
Self-updating knowledge per run
Plug-in layer for LangGraph, CrewAI

Performance Marketing

Every brief knows every campaign.

Creative briefs, competitor ads, winning hooks, brand guardrails. Stored as vectors, linked as a graph. One query surfaces the full campaign memory instantly.

Multimodal: copy + creative + video
Brand-safe context per namespace
Hawky.ai native integration

Enterprise AI

The context layer your LLM stack is missing.

Wikis, specs, calls, tickets. The private knowledge that makes your business yours. Feather keeps it fresh, filtered, and sub-millisecond to retrieve.

Multi-tenant per workspace
Deploy in your VPC (Cloud tier)
Role-based metadata filters

Developer Tools

Memory for the tools that build software.

IDE assistants, repo-aware agents, autonomous workflows. Feather's embedded mode drops into any toolchain. No server, no network hop, just a file.

Embedded in-process
Zero infra for CLI tools
Works offline, syncs when online

Community

What builders are saying.

Shipped in early preview. Open source since day one. Here's what the community has to say.

“Context_chain replaced 400 lines of our retrieval+rerank code. One call, and the agent has everything it needs.”

Lea M.Staff Eng, autonomous agents

“Feather is weirdly fast. Sub-millisecond at 100k vectors without tuning anything. The C++ core is doing real work.”

@vectornerdGitHub

“MIT license. C++ core. Python bindings. Rust CLI. It's every box ticked and then some.”

Tomás B.OSS maintainer

“The adaptive decay is the piece every other vector DB is missing. Our memory actually stays relevant week to week.”

Daniel K.Founder, AI copilot startup

“We ripped out Pinecone for local-first development. Ship speed is 3x.”

Ana S.Head of AI, fintech

“I expected an early-stage OSS project. I got a production engine with clean APIs and benchmarks that hold up.”

Marcus W.Platform eng, enterprise SaaS

“A single .feather file on disk. No server, no container. For our edge deployments this is genuinely the only thing that works.”

Priya R.Infra lead

“The graph + vector unification is the right mental model. I stopped maintaining two stores.”

@davidzResearcher

“Hawky.ai's creative memory runs on Feather. It's the core of why our agents know what they're doing.”

Ashwath R.Founder, Hawky.ai

Pricing

Simple, transparent pricing.

Start free with Core. Pay only when you're at scale. No seat taxes, no surprises.

Feather Core

For solo devs, OSS, edge

Free foreverSelf-hosted · Community support

Embedded, single .feather file
Python + Rust SDK, CLI
Semantic + graph + metadata
BM25 + hybrid RRF search
SIMD AVX2/AVX512 core
MIT license · Community support

Install now

Most loved

Feather Cloud

For teams scaling up

Usage-based pay-as-you-goManaged · Priority support

Everything in Core
Managed API
Horizontal auto-scale
VPC deployment option
Priority support
99.9% SLA

Enterprise

For regulated & large scale

Custom tailored to youVPC / On-prem · Dedicated support

Everything in Cloud
On-prem or VPC
Custom SLAs
Dedicated engineer
Security review & SOC2
Training & migration

Open source under MIT. Your .feather file is yours, forever.

Give your agents a memory worth keeping.

Install Feather Core in 5 minutes, or talk to us about the managed Cloud edition.

View on GitHub