Your AI agent can now pay per call (x402, live in v0.19)
Your agents already forget. Stop letting them.
AllSource records every event your agent emits, with 12μs recall on every restart. Pay only when the agent reads.
Open source · MIT licensed · Self-host or cloud
What did the user do yesterday at 3pm?
recalling from event log…
The Problem
AI Integration Gap
Your data is trapped behind rigid APIs. AI agents can't easily explore, analyze, or manage your event streams. The tools that exist weren't designed for autonomous workflows — your agents forget everything between sessions.
State Amnesia
Traditional databases only store current state. When something goes wrong, you can't see how you got there. No history means no answers, no audit trail, and no ability to replay or debug.
Temporal Blindness
Time-based queries are afterthoughts. Want to know what your data looked like last Tuesday at 3pm? Good luck. Most systems weren't built for time-travel, and retrofitting is painful.
The Solution
AllSource combines high-performance event sourcing with AI-native tooling. Rust for speed, Go for orchestration, Elixir for distributed queries. The right tool for each job.
Immutable Event Log
Every change is captured as an immutable event. Complete audit trails, perfect reproducibility, and the ability to reconstruct any past state with time-travel queries.
Sub-Microsecond Queries
11.9μs p99 query latency with Parquet columnar storage. Lock-free data structures and zero-cost field access deliver real-time analytics at scale.
AI-Native by Design
43-tool MCP Server for Claude Desktop. AI agents can manage events, run analytics, detect anomalies, and orchestrate complex workflows autonomously.
Stream Processing Pipelines
Filter, Map, Reduce, Window, Branch, Enrich. Build real-time pipelines with a fluent API. Process 469K events/sec with the Rust core engine.
How It Works
Time-travel your data.
Why AllSource
Event sourcing infrastructure with an optional AI agent memory module (Prime). The core event store stands on its own — Prime adds knowledge graphs and vector search for teams that need it.
| Feature | AllSource | mem0 | Letta | Zep |
|---|---|---|---|---|
| Temporal queries (as_of) | ||||
| Full event provenance | ||||
| Auto compressed index | ||||
| Offline / embedded mode | ||||
| Sub-millisecond recall | ||||
| MIT licensed / self-host |
Based on public documentation and independent benchmarks. Last updated April 2026.
Event Intelligence
Built in Rust for performance. Designed for AI agents. Every feature accessible through 43 MCP tools — let Claude or GPT manage your event streams autonomously.
Time-Travel Queries
Reconstruct any entity's state at any point in time with as_of queries. Debug issues, audit changes, or replay historical states instantly.
Microsecond Latency
11.9μs p99 query latency with Parquet-based storage. Optimized for both high-throughput ingestion and fast analytical queries.
Event Replay
Rebuild projections, fix incorrect state, or migrate data by replaying events. Filter by entity, type, or time range.
Projections
Materialized views that update automatically. Entity snapshots, counters, time series, funnels, and custom projections.
Schema Registry
Versioned schemas with backward, forward, and full compatibility modes. Validate events at ingestion time.
Multi-Tenant Isolation
Every event, projection, and query is scoped to a tenant. RBAC with 4 roles and configurable quotas per tier.
Event Sourcing
- Immutable audit logs
- State reconstruction
- Event replay
AI Workflows
- 43 MCP tools
- Claude integration
- Autonomous agents
Analytics
- Stream processing
- Real-time pipelines
- Correlation analysis
Pricing
Self-Host
Free
your infra
- Unlimited events (your hardware)
- Forever retention
- Unlimited streams
- Full MCP (self-host)
- GitHub community support
- MIT licensed
Indie
$18.99/mo
billed monthly
Includes 50K x402 calls. $0.0001/call after.
- 500K events/month
- 14-day retention
- 3 streams
- Hosted MCP: read
- Email support (48h)
Studio
$78.99/mo
billed monthly
Includes 500K x402 calls. $0.0001/call after.
- 5M events/month
- 90-day retention
- Unlimited streams
- Hosted MCP: read + write
- Email support (24h) + Discord
Scale
$298.99/mo
billed monthly
Includes 5M x402 calls. $0.0001/call after.
- 50M events/month
- 365-day retention
- Unlimited streams
- Hosted MCP: read + write + dedicated
- Priority support + Slack
Enterprise
Regulated / SLA workloads. Dedicated MCP cluster. Negotiated volume, unlimited retention, 24/7 + dedicated SE.
FAQ
Blog
Prime Stopped Being a Database
Our hosted agent-memory engine used to run its own store. We made it stateless — it now reads and writes Core over HTTP, isolated per tenant, with no durable storage of its own. Here's the architecture, the load-bearing trick, and the bug we caught in production.
Quantized Index vs Durable Memory: turbovec and AllSource Prime
turbovec compresses vectors 8–16× with no recall tax. AllSource Prime makes vector memory durable, graph-aware, and replayable. They aren't rivals — they own different layers of the same stack, and they compose. Here's the whole stack, drawn out, with the use cases for each and for both together.
chronis 0.8.0: the web viewer grows up
chronis 0.8.0 turns `cn serve` from a read-only list into a real workspace viewer — sortable dashboard columns, an expand-in-place detail pane with two-way dependencies, a force-directed bubble graph, and a cleaned-up tree. All offline, all served from one Rust binary.
Why Your Agent's Memory Returned Nothing — and How We Fixed Hybrid Recall
AllSource Prime's compressed index reported zero nodes for everything you recorded, and prime_context's L2 tier returned no vectors. Three bugs, one event-sourcing lesson: a projection that isn't fed is a projection that lies. Here's the debugging story, the fix, and the real before/after numbers from the binary.
Your Voice File Shouldn't Be a Dead Markdown File
The viral 'voice file' is right: in 2026 your identity layer is the moat, not your prompts. But a static 4k-token markdown blob goes stale, can't be queried, and has no history. prime makes your voice live — record facets once, recall the relevant slice by meaning per task. Proven with a real voice-ON vs voice-OFF demo.
Your Coding Agent Has Goldfish Memory. Give It a Mammoth.
mammoth gives any coding agent durable, cross-session memory — decisions, task history, and code context recalled automatically, so you stop re-explaining your project. Local-first, zero signup, MCP-native. Here's how it works and what the benchmarks say.
Using AllSource as a CMS from Claude Desktop
An agent-executable install protocol for AllSource Prime, plus the working pitch: your CMS already lives where your agents work. Written so a Claude Desktop agent can read this article on a human's behalf and drive the install with a bounded, two-action human role.
Prime 0.21.4: text-only embeds, sync, and a Memory tab
Three shipping changes for AllSource Prime — prime_embed and prime_recall take plain text, prime-mcp can push events to your tenant Core, and the panel grew a Memory tab to show them. Install in 30 seconds.
An Agent That Provisions Its Own Persistence in One API Call
Your agent doesn't need a human to set up storage. One POST to AllSource and it has a tenant, an API key, and 100K events waiting to be written.
Write Before You Execute: Building Crash-Safe AI Agents
An agent that crashes mid-task and then re-runs can send the same email twice, create duplicate orders, or push the same commit twice. Here's a single pattern that prevents all of it.
How Agent Teams Stay in Sync with a Shared Event Stream
When multiple AI agents work on the same problem, they need a shared world model — not direct calls to each other. Here's how to build it with AllSource and chronis.
AllSource for Startups: From Local Dev to Production in 15 Minutes
The fastest path from zero to event sourcing for startups and side projects. Self-service onboard, free tier, no infrastructure to manage. Move to production when you're ready — not before.
How AllSource Core Works: WAL, Parquet, and DashMap
A deep-dive into AllSource Core's storage architecture. Write-Ahead Log with CRC32 checksums, Parquet columnar persistence, and DashMap concurrent reads. How we get 469K events/sec with zero data loss.
How to Build Audit Trails That Pass SOC2 with Event Sourcing
A practical guide to building SOC2-compliant audit trails using AllSource event sourcing. Immutable event logs, time-travel queries, and cryptographic integrity — without the compliance tax.
Event Sourcing for AI Agent Memory: A Practical Guide
How to use AllSource as durable, time-travelling memory for LLM agents. Store conversations, decisions, and context as events. Query any past state. Give agents memory that survives restarts.
Real-Time Dashboards Without ETL: Event Sourcing + Projections
How projections replace traditional ETL pipelines for real-time dashboards. AllSource event sourcing keeps materialized views in sync as events arrive — no Kafka, no Airflow, no batch jobs.
Why Your SaaS Needs Event Sourcing (Not Just a Database)
The business case for event sourcing in multi-tenant SaaS. Audit trails, tenant isolation, usage billing, and feature flags — all as a natural consequence of storing events instead of state.
Connecting to AllSource without an SDK
How to ingest, query, stream, and build projections against AllSource using only HTTP and WebSockets — no SDK required. For teams on unsupported languages or anyone who wants to understand the wire protocol.
Direct to Core, or through the gateway? Choosing your connection path
AllSource has one public front door (api.all-source.xyz) and one internal fast path (Core, reachable only inside your network). Picking the right one cuts your p99 in half — picking the wrong one either reimplements rate limits or exposes your event store to the internet.
Tiered Context Loading: Cut Agent Memory Costs by 60% Without Losing Recall
Most agent turns don't need full recall. L0/L1/L2 tiers let agent loops fetch exactly the context depth they need — from 100-token stats to full hybrid retrieval — reducing token costs while keeping accuracy where it matters.
12μs Agent Memory: How We Got There
From event store to agent memory engine in microseconds. DashMap projections, HNSW vector index, and why the storage layer matters more than the query layer.
Building Agent Memory in Rust: From Event Store to Knowledge Graph
How we built a unified agent memory engine — vectors + graph + compressed index — on top of an event store. Architecture decisions, SOLID refactoring, and why Rust was the right choice.
How a Compressed Index Doubles Cross-Domain Recall
Vector similarity finds X or Y — rarely both. A 500-token markdown index bridges the gap. Here's why, with benchmark data.
From zer0dex to AllSource: What We Learned
We analyzed zer0dex's dual-layer memory system, found the insight that matters, and built on it. Here's what we kept, what we changed, and why.
zer0dex vs AllSource: What Agent Memory Actually Needs
zer0dex proved that a compressed index doubles cross-domain recall. We took that idea, made it automatic, and added temporal reasoning. Here's the honest comparison.
Your AI Agents Need Memory, Not Just Storage
AI agents are getting smarter, but they still can't remember yesterday. Here's why temporal context is the missing piece in agentic AI.
Event Store vs Database: Choosing the Right Foundation
Databases store state. Event stores store history. Here's how to decide which foundation your application actually needs.
Temporal AI: Why Your RAG Pipeline Needs a Timeline
Vector search finds similar content. Event sourcing tracks history. Combined, they create AI that truly understands context.
Why Event Sourcing in 2026: Beyond Simple Storage
Your application needs more than a database—it needs perfect memory. Here's why event sourcing matters more than ever.
AI Agents with MCP Tools
How to let Claude and other AI agents manage your event streams autonomously.
Time-Travel Queries Explained
How to reconstruct any entity's state at any point in time with all.source.
Introducing all.source
Time-travel your data with the AI-native event store built for the future.
Give your AI agents an event log that remembers every event
No credit card required. 10K events/month free.