Contextual
Paid
Individual developers managing local codebases
Memori
FreemiumAPISelf-HostedAgentic
Every agent that forgets what it did last session forces you to either bloat the prompt with full context or watch it repeat the same mistakes — and at scale, both options cost real money. Memori is a memory infrastructure layer that captures agent execution traces and conversation turns, structures them, and retrieves only what's relevant when the next prompt fires.
Similar tools
ContextualPaidThe engine indexes your repository — symbols, commits, PR threads, docs — into a local graph…
AI·rete·RAGFreemiumThe tool pairs a deterministic rule engine with retrieval-augmented generation, so every…
JylusFreemiumJylus ingests streaming data, resolves current and historical state, follows relationships…
SteleFreemiumStele is a shared memory layer that sits between your agents and your codebase. Every agent…
Screenshots 2
Summary
The vendor states Memori classifies each chat turn into facts, preferences, rules, and summaries, then pulls targeted snippets at recall time rather than re-injecting full history. On the LoCoMo benchmark, the docs report 81.95% accuracy while cutting token usage by 95% versus full-context retrieval — a meaningful number if your cost problem is upstream of the model choice. The memory graph shows how entities connect across sessions, and every recall result ships with lineage explaining why that snippet was included, which matters when an enterprise audit asks why the agent said what it said. The ceiling appears when your retrieval logic needs fine-grained control the SDK's zero-configuration defaults don't expose — teams at that point are writing wrapper logic to compensate. Self-hosted deployment is available, so organizations with data-residency requirements are not locked into the cloud path.
Bottom line: Pick Memori when your production agent is bleeding tokens replaying full conversation history and you need auditable recall fast — but plan for additional engineering if your multi-hop retrieval accuracy needs to clear the 72.70% the benchmark reports for that query type.
Pricing Plans
Flat RateLast verified 4 months ago- Price
- $19/month
- Free Tier
- 5,000 memories created, 15,000 memories recalled
Free
Free
For builders exploring Memori or running lightweight workflows.
- 5,000 memories created
- 15,000 memories recalled
- Advanced Augmentation
- Unlimited end users
- 30-day usage reset
- Community Support
Starter
$19per month
For growing apps, agents, or prototypes ready to step into production territory.
- 25,000 memories created
- 100,000 memories recalled
- Advanced Augmentation
- Unlimited end users
- 30-day usage reset
- Community Support
Pro
$99per month
A strong fit for teams running consistent production loads, customer-facing apps, or multi-agent systems.
- 150,000 memories created
- 500,000 memories recalled
- Advanced Augmentation
- Unlimited end users
- 30-day usage reset
- Private Slack Channel
- Ability to purchase add-on packs
Enterprise
Custom
Custom plans. Forward-deployed engineers. Dedicated integration and maintenance support.
- Forward-deployed engineers
- Dedicated integration and maintenance support
View full pricing on memorilabs.ai →
Pricing may have changed since last verified. Check the official site for current plans.
Find your fit
Before you choose Memori
More options in RAG Frameworks. Compare pricing, platforms, and features side by side.
AI·rete·RAG
Freemium
Teams needing deterministic rules plus RAG explanations
Pros
- Classifies memory into typed categories (facts, preferences, rules, summaries) at write time, so recall is targeted rather than probabilistic — which means your agent isn't paying token costs to re-read irrelevant history on every turn.
- The vendor reports 95% token reduction versus full-context retrieval on the LoCoMo benchmark, so teams with high-volume agents stop absorbing LLM spend just to maintain conversational continuity.
- Every recall result includes lineage tracing the entity, time, and source of inclusion, so when an enterprise audit asks why the agent surfaced a specific piece of context, there is a concrete answer rather than an opaque embedding distance.
- LLM-agnostic architecture means switching the underlying model — from OpenAI to a self-hosted alternative, for example — does not force a memory layer rewrite.
- Self-hosted deployment is available, so teams with data-residency or compliance requirements are not forced onto the cloud path.
Cons
- Multi-hop recall accuracy benchmarks at 72.70% and open-domain at 63.54% — agents that chain several inferential steps across memory or handle unconstrained queries will surface wrong context at a measurable rate, and teams building those workflows are adding custom retrieval logic on top, at which point they are maintaining two systems.
- The zero-configuration SDK default is fast to ship but exposes precious little surface area for teams that need fine-grained control over retrieval scoring, memory expiry policies, or scoping rules beyond what the defaults provide — those teams end up writing wrapper logic that grows in complexity as production edge cases accumulate.
- Closed-source with no self-service inspection of the classification or recall logic means when the memory layer returns unexpected results, debugging is limited to the lineage output the tool surfaces — teams that need to audit or modify the core retrieval behavior switch to an open-source alternative they can instrument directly.
About
- Platforms
- Cloud (Memori Cloud), Self-hosted via open-source SDK
- API Available
- Yes
- Self-Hosted
- Yes
- Last Updated
- June 9, 2026
Best For
Who it's for
- Teams building production AI agents with long-running workflows
- Applications requiring persistent memory across user sessions
- Systems optimizing inference costs through structured context
- Enterprise deployments needing auditable memory state
What it does well
- Customer support agents with multi-session memory
- Commerce automation workflows that learn from prior execution
- Internal copilots that remember organizational context
- Multi-agent systems requiring shared, scoped memory
- Production robotics requiring persistent personalization
Integrations
Compare Memori
Frequently Asked Questions
- Is Memori free?
- Memori offers a free option alongside paid upgrades. Pricing: $19/month. Check the current allowances and terms before choosing a plan.
- Does Memori have an API?
- Yes. Memori exposes a developer API. See the official documentation at https://memorilabs.ai for details.
- Can I self-host Memori?
- Yes. Memori supports self-hosting on your own infrastructure.
- When was Memori released?
- Memori was first released in 2024.
- What platforms does Memori support?
- Memori is available on: Cloud (Memori Cloud), Self-hosted via open-source SDK.
People also compared
Curated lists that include this category
Agents that run across sessions accumulate context — and without a structured way to store and recall it, you are either stuffing the prompt until it’s expensive, or starting fresh until it’s useless. Memori sits between your agent and the LLM as a memory layer: it intercepts execution traces and conversation turns, classifies them into typed memory (facts, preferences, rules, summaries), stores them, and at inference time retrieves only the slice of context relevant to the current prompt. The SDK drops into existing code with, as the vendor describes, zero configuration for model calls and callbacks.
The differentiating architectural choice is structured recall over retrieval. Rather than vector search across raw conversation dumps, Memori classifies memory at write time and enriches searches with semantic context at read time — the vendor describes this as ‘selective semantic search’ that improves accuracy without expanding token payloads. The LoCoMo benchmark results the vendor publishes show 81.95% overall accuracy with a 95% reduction in token usage versus full-context approaches. Every recall result includes lineage: entity, time, and source of why that memory was included, which gives engineering and compliance teams a traceable audit trail instead of a black-box retrieval.
Memori fits best in production systems where agents run long, stateful workflows — customer support agents that need to remember a user’s prior issues, commerce automation that learns from prior execution paths, internal copilots that accumulate organizational knowledge over time. Where it shows strain: multi-hop queries benchmark at 72.70% and open-domain at 63.54%, meaning agents that need to chain several inferential steps across memory or handle broad, unconstrained queries will produce less reliable recall. Teams hitting that ceiling are adding their own retrieval layers on top, which means maintaining two systems. The self-hosted option addresses enterprise data-residency requirements, and the API makes integration with existing agent frameworks direct — but advanced recall configuration beyond the defaults requires digging into the SDK rather than a visual interface.
Contribute
- Work on Memori? Claim this listing.
- Embed a badge for your README or site.
Sign in to rate this tool or suggest an edit.