# Alternatives to Letta > Source: https://theagentsindex.com/letta/alternatives (current datasheets, evidence-ranked) Candidates share the baseline category or are previously researched peers. Requirements, preserved strengths, explicit trade-offs and research coverage decide the order; paid tier does not. ## Shortlist ### 1. OpenCode Open source AI coding agent for the terminal, desktop, browser and IDE, running any model you connect. **Research coverage:** 33/35 criteria **Where it improves on Letta:** - [Pooled budget: yes — A Zen workspace draws on one balance with a workspace-wide monthly cap plus optional per-member caps.](https://opencode.ai/docs/zen) - [Admin controls: yes — Permissions are set in config globally or per tool, so an operator can deny edits or shell commands outright.](https://opencode.ai/docs/permissions) **Trade-offs against Letta:** - No researched regression against the baseline. [View OpenCode](https://theagentsindex.com/opencode) ### 2. OpenHands Open-source software engineering agent with a browser client, CLI and Python SDK; local, cloud or self-hosted. **Research coverage:** 34/35 criteria **Where it improves on Letta:** - [Writes tests: yes — Test generation and coverage expansion is a named workflow; a separate QA agent also exercises the running software rather than the test suite.](https://www.openhands.dev/services) - [MCP server: yes — Agent Server exposes an OpenAI-compatible chat-completions endpoint that another agent or client can call, holding conversation state behind a returned header.](https://docs.openhands.dev/sdk/guides/agent-server/openai-gateway) - [Pooled budget: partial — Credits are pooled across all members, but only inside an Organization, which requires a paid Cloud subscription or Enterprise.](https://docs.openhands.dev/openhands/usage/cloud/organizations/overview) **Trade-offs against Letta:** - [In your editor: partial — Zed, JetBrains and Toad connect natively over the Agent Client Protocol and VS Code via a community extension, but the vendor labels the whole integration experimental.](https://docs.openhands.dev/openhands/usage/cli/ide/overview) - [Seat model: no — Enterprise is explicitly not licensed per seat and carries unlimited users; the two free tiers are each capped at one user.](https://docs.openhands.dev/enterprise) [View OpenHands](https://theagentsindex.com/openhands) ### 3. Agno Python SDK, open-source AgentOS runtime and hosted control plane for building and operating agents. **Research coverage:** 35/35 criteria **Where it improves on Letta:** - [MCP server: yes — AgentOS can expose registered agents as an MCP server, so another agent can call them; docs also ship as MCP.](https://docs.agno.com/features/api) - [Seat model: yes — Three seats included on Pro, $30/month per extra seat; no published ceiling.](https://www.agno.com/pricing) - [Pooled budget: partial — There is no metered balance to pool; seats share access to the same connected runtimes and usage inside a connection is unlimited.](https://www.agno.com/pricing) **Trade-offs against Letta:** - [Agent permissions: partial — Runs are unattended unless a tool is marked for approval; then the run pauses in your database until an admin resolves it.](https://docs.agno.com/agent-os/approvals/overview) - [In your pipeline: no — The enumerated serving surfaces are REST, MCP in Claude/ChatGPT and chat apps; no pull-request or CI surface appears.](https://www.agno.com/) - [Open source: partial — SDK and AgentOS are Apache 2.0; the control plane is the commercial part and is not published as source.](https://www.agno.com/compare/agno-vs-langgraph) [View Agno](https://theagentsindex.com/agno) ### 4. Cursor Coding agent in a desktop editor, terminal CLI, cloud VMs and CI, priced per seat plus token usage. **Research coverage:** 35/35 criteria **Where it improves on Letta:** - [Writes tests: yes — Test generation is named as agent work, and Automations can be configured to add tests on every new commit.](https://cursor.com/automate) - [Whole-codebase work: yes — Cloud agents work across multiple repositories in one task and raise a pull request in each repo they touch.](https://cursor.com/docs/cloud-agent) - [MCP server: partial — MCP is inbound only: Cursor is the client that connects to MCP servers. Another agent drives Cursor over ACP (agent acp) or the SDKs, not by querying an MCP endpoint Cursor exposes.](https://cursor.com/docs/mcp) **Trade-offs against Letta:** - [Open models: partial — All inference is hosted: the catalogue includes GLM 5.2 from Z.ai and Kimi K3 from Moonshot, but there is no path to point Cursor at open weights you run.](https://cursor.com/docs/models-and-pricing) - [Self-hosted: partial — Only tool execution moves to your machines: Cursor still runs the agent loop, inference and planning, and the Self-Hosted Pool needs Enterprise. The Enterprise FAQ says there is no on-premises deployment.](https://cursor.com/docs/cloud-agent/self-hosted) - [Open source: no — The Terms of Service forbid deriving or accessing the source; the /licenses page covers only third-party open source components used inside the IDE.](https://cursor.com/terms-of-service) [View Cursor](https://theagentsindex.com/cursor) ### 5. Codex CLI OpenAI's terminal coding agent for inspecting, editing and running repository code from the shell. **Research coverage:** 35/35 criteria **Where it improves on Letta:** - [Writes tests: yes — Documented CLI workflow for writing a unit test for a named function.](https://learn.chatgpt.com/docs/prompting) - [MCP server: partial — Codex is an MCP client for local and remote servers, but exposing Codex itself as an MCP server is deprecated in favour of the app server.](https://learn.chatgpt.com/docs/codex-sdk) - [Seat model: yes — Per-user pricing with a two-seat floor and no published ceiling; monthly billing costs $25 per user.](https://learn.chatgpt.com/docs/pricing) **Trade-offs against Letta:** - [Claude: no — Sign-in and model support are OpenAI models only; no Anthropic model appears in any documented model list.](https://learn.chatgpt.com/docs/auth) - [Model choice: partial — You pick the model per run and can define custom providers, but the first-party catalogue is OpenAI models only.](https://learn.chatgpt.com/docs/config-file/config-advanced) - [Self-hosted: partial — The agent binary runs on your machine or runner; inference is OpenAI-hosted unless you route it through your own AWS Bedrock account.](https://learn.chatgpt.com/docs/amazon-bedrock) [View Codex CLI](https://theagentsindex.com/codex-cli) ### 6. Factory Droid coding agents that run in the desktop app, CLI, IDEs, CI and Factory-managed cloud computers. **Research coverage:** 35/35 criteria **Where it improves on Letta:** - [Writes tests: yes — The /install-qa skill generates QA sub-skills and executes end-to-end test flows per diff, with evidence.](https://docs.factory.ai/software-factory/automated-qa) - [Whole-codebase work: yes — Missions decompose multi-feature work into milestones and are documented for brownfield migrations.](https://docs.factory.ai/missions/overview) - [MCP server: partial — Droid is an MCP client over stdio/http/sse and can be driven by another program over JSON-RPC or ACP, but no Factory-hosted MCP server is documented.](https://docs.factory.ai/harness/mcp) **Trade-offs against Letta:** - [Self-hosted: partial — The agent always runs on your machine, and a dedicated airgap build exists, but airgap builds are Enterprise-only and not publicly distributed.](https://docs.factory.ai/enterprise/airgapped-deployment) - [Open source: partial — The CI action is a public repository; the Droid agent itself ships as binaries, with no source offered.](https://docs.factory.ai/software-factory/code-review-ci) - [Free tier: no — The published plan list starts at Pro $20/mo; no free tier or trial is offered, though Droid Core and a BYOK allowance are free within a paid plan.](https://docs.factory.ai/pricing/individuals) [View Factory](https://theagentsindex.com/factory-ai) ## Method Unknown data is never treated as a benefit or a trade-off. Featured placement is labelled but cannot alter recommendation order.