Letta
by Letta · Stateful AI agents with persistent memory, via CLI, desktop and web
BenchRank score
Screenshots of Letta
Homepage
Overview
Letta is a San Francisco AI research lab whose product, Letta Agent, is a stateful agent whose memory, identity and capabilities change with use. It runs through a CLI installed with npm, plus a desktop app, web app and API, and uses either Letta Auto model routing or your own API keys. Agents can be stored locally or self-hosted, and shared across an organisation.
- Best for
- Individuals and teams running stateful AI agents for coding and chat, via CLI or desktop app
- Pricing
- Free plan at $0/month with bring-your-own-key models, then Pro at $20/month; developer plans are usage-based and Teams/Enterprise prices are not shown.
- Runs on
- Self-hostedCLI
Strengths and trade-offs
Strengths
- Free plan runs on your own API keys and coding plans
- Pro is $20/month and adds remote sandboxes, 20 agents
- Agents can be shared across a team with access permissions
- Usable as CLI, desktop app, web app and API
Trade-offs
- Free plan is capped at 3 stateful agents, Pro at 20
- Coding-heavy use is quoted at ~$100-$200+/month in total usage
- Docs advise pairing it with a third-party coding plan for heavy use
- Teams Pro and Developer plan prices are not published on the page
How Letta markets itself
A structured read of the promise, proof and page design on Letta’s captured homepage.
Homepage capture
“Letta is an AI research lab in San Francisco building machines that learn. We envision a future where humans coexist with digital people: experiential agents that remember everything, learn continuously, and improve themselves over time.”
- Angle: Authority / credibility
- Hero: 3D render
Pricing
Published plans and prices from Letta’s own pricing page.
How this score is made up
Each dimension is scored out of 100 and combined into the headline score using fixed weights.
MCP support
Whether an agent can drive the product through the Model Context Protocol, and how much setup that takes.
API quality
Public API surface: machine-readable spec, official SDKs, documented auth, errors, rate limits and versioning.
Documentation
Publicly reachable docs — coverage, freshness, code samples and machine readability.
Agent friendliness
How readable the site is to an automated client: llms.txt, structured data, server-rendered content, crawler access.
Pricing transparency
Whether real prices are published, self-serve signup exists, and usage costs are knowable without a sales call.
Changelog
A public, dated record of what shipped and when — the clearest signal that a product is still alive.
Marketing site structure
Whether the site answers a buyer's questions: clear positioning, the pages that matter, and accessibility.
Page speed
How fast the site loads for real visitors: Chrome UX Report 75th-percentile LCP, INP and CLS, with a Lighthouse mobile run standing in where a site has too little traffic for field data.
Operational trust
Status page and incident history, security disclosure, compliance and data-processing documentation.
Measured, but not part of the score
Useful to know, but not a mark for or against the product — so these do not affect the ranking.
Openness
Source availability, self-hosting, data export and open standards. Scored and shown, but not part of the composite — paid SaaS is not worse for being paid SaaS.
Maintenance
Release cadence and repository activity. Scored and shown, but not part of the composite — it is only measurable for open repositories.
This doesn’t look right — report a problem with Letta’s score
Where this comes from
The Letta pages BenchRank reads when it scores the product — its documentation, release notes, status and security pages, and its repository where there is one.
Alternatives in AI Development Platforms
Ranked 1
85.5 — BenchRank score out of 100Mem0
Mem0 · Hosted memory layer for AI agents, with Python and Node SDKs
Best for: Developer teams adding persistent memory to AI agents through a hosted API, with a free tier
Ranked 2
83.7 — BenchRank score out of 100Nango
Nango · Code-first integration platform covering 900+ APIs
Best for: Product teams building many third-party API integrations into a SaaS product or AI agent
Ranked 3
81.4 — BenchRank score out of 100Supermemory
Supermemory · Memory and retrieval layer for AI agents
Best for: Developers giving AI agents persistent memory and retrieval through a single hosted API.
Is this your product?
Claim Letta to manage its profile. Claiming lets you suggest edits to the descriptive fields — it never changes scores or rankings.