lemma-platform
Lemma is an open-source runtime for building and running agent-built software your whole team shares.
Lemma earns its keep when an agentic workflow touches money, access, or customer records and more than one person needs to use it. The pod model — shared running agent, one set of records, permissions that bind humans and agents identically — is the differentiator, not the chat surfaces. Reach for it over hand-assembling CrewAI or LangGraph when you need logged approvals and per-table grants rather than a demo. Go in expecting a CLI-and-coding setup loop, not a no-code builder.
Verified 7d ago · liveness 73/100 · cite: rightaichoice.com/tools/lemma-platform
- Engineering teams building internal agentic workflows with human-in-the-loop approvals
- Ops teams automating refunds, access requests, or incident coordination across chat surfaces
- Developers authoring pods as plain files with Claude Code, Codex, Cursor, or OpenCode
- Teams that need self-hosted agent tooling with per-table RBAC and logged approvals
- Non-technical users who want a no-code AI assistant without touching a CLI
- Teams wanting a broad catalog of pre-built enterprise integrations out of the box
- Anyone looking for a consumer chatbot rather than a deployed internal tool
We scan live Reddit threads, YouTube comments, X posts, G2 reviews and other communities — and hand you an honest verdict in under a minute.
- Honest verdict, not marketing
- Real pros & cons from real users
- Attributed quotes with receipts
3 free scans · no card needed
Skip Lemma if nobody on your team is comfortable authoring plain files through a coding agent or the CLI — this is a developer runtime for deployed internal tools, not a no-code assistant you configure in a browser.
Choosing your own provider key means model usage is billed by that provider on top of the runtime — Lemma doesn't absorb those inference costs.
The open-source runtime is free to self-host, so the real cost comparison is your own operational time plus whatever your model provider charges. That puts Lemma in a different bracket from managed no-code agent builders with per-task pricing, and often cheaper than a managed equivalent at the same headcount — the trade is that you supply the setup labor and the hosting.
In short
lemma-platform — Lemma is an open-source runtime for building and running agent-built software your whole team shares. Best for Engineering teams building internal agentic workflows with human-in-the-loop approvals, Ops teams automating refunds, access requests, or incident coordination across chat surfaces, Developers authoring pods as plain files with Claude Code, Codex, Cursor, or OpenCode. Free to use.
What's new in lemma-platform
Checked 7 days agoAcross the latest 5 updates: 1 launch and 4 changelog entries.
September 24 — The new workspace and clearer desktop setup
First-start workspace downloads show progress when size is known, Codex tool cards show clearer commands, output and failure details, and host agents are directed to the in-app sandbox browser for browser work.
Teammate workspace merged into the platform repository
Conversations, apps, files and teammate profiles now share one interface. The operator application remains available as Lemma Harness, and desktop builds ship through the September 24 nightly.
September 21 — Shared channels and browser handoff
Slack bot routing identifies the surface that owns a conversation and threaded replies are handled only where the teammate participates. Browser pane gains human takeover and paste support.
September 16 — Editable widgets and a shared browser
Widgets can live as files the agent reopens and edits, interactive widgets hand follow-up questions back to the conversation, and sandbox work gains a browser surface, file access and saved browser logins.
0.8.0 — Three names on the wire settle on one spelling; Lemma runs on Windows
Three HTTP API path/field renames land ({org_id} → {organization_id}, pods nested under organizations, colon action routes become slash routes), and Lemma runs on Windows for the first time.
What people actually say about lemma-platform — is it worth it?
We ran a structured research pass across product reviews, community discussions, and post-purchase forum threads to surface the patterns vendors won't publish themselves. Below: the recurring strengths, the hidden costs people mention most, and the cohort that consistently regrets adopting this tool.
14 mentions across 2 sources (YouTube, GitHub) · researched Jul 1, 2026.
Average across the 2 sources that answered — each source counts once, not each post.
- +Open-source workspace for humans and AI agents working as one team.
- +Pods combine data, agents, workflows, and human interfaces in one environment.
- +Conversation-first agent interactions with streaming and metadata support.
- +Granular access scoping and approval workflows for shared pods.
- +SDK support includes Core Client, React hooks, and authentication.
- −Extremely sparse community data — mostly game reviews, not platform feedback.
- −Bug: agent 'pod_write_record' strips payloads to empty, causing write errors.
- −Bug: stale CLI tool names and sonnet default to 1M context without credits.
- −Windows compatibility broken: 'ModuleNotFoundError: No module named termios'.
- −Missing webhook/callback support for run completion notifications.
- • Cloud service costs if hosted on own infrastructure
- • Potential cost for running large language models via external APIs
Viability Score
How well maintained and how widely used is lemma-platform? Built from what the vendor actually publishes (docs, changelog, tutorials, integrations, pricing), whether the site is live, and how much real users discuss it. How we calculate this
Last calculated: October 2026
How we score →Key Features
- Pod-based architecture bundling tables, files, functions, agents, workflows, and app
- Slack as a native surface with streaming replies and App Home setup (0.7.0)
- Microsoft Teams as a native surface
- Desktop app for macOS, Linux, and Windows (0.7.1 onward; Windows host pack added in 0.8.0)
- Agent memory: /memory for pod-shared facts, /me for personal context
- Model selection popover with recent picks and full catalog
- Bring your own provider key, run Lemma-hosted models, or pair a local coding agent
- Agents run on schedules, webhooks, table events, messages, or the API
- Serverless functions for deterministic logic, callable by agents as tools
- CLI for authoring and importing pods as plain files, with verification
- Versioned and rollback-able deployments behind a single URL
- Per-table grants, resource visibility, and approval gates for humans and agents
- Row-level security on tables plus full-text searchable markdown policy files
- Workflows that pause at a human approval and resume on their decision
- In-app sandbox browser with saved logins and browser recordings
About lemma-platform
Lemma is an open-source runtime for agent-built software: multi-agent applications that a team shares as one running instance rather than a pile of copies. Engineering and ops teams use it to stand up internal tools — incident coordinators, approval bots, AI inboxes, account consoles, release operators — and reach them from Slack, Microsoft Teams, Telegram, WhatsApp, Gmail, or a custom web app. Chat is the surface; the pod is the system. The core unit is the pod: one boundary wrapping tables and files, serverless functions, agents and workflows, the app people open, and the permissions layer underneath. Records come first, then deterministic functions, then the agents and workflows that move work. Because you share the agent itself, already running, there is one instance to fix, one set of records beneath it, and one version for everyone. Access control is the part most agent platforms skip. People and agents are both members of the pod and follow the same rules — per-table grants, hidden resources, approval gates on risky steps. A classifier agent can read tickets and write status while being blocked from refunds, API keys, and connectors. Workflows pause at a human approval and resume on that person's decision, days later if needed. Pods are authored as plain files by coding agents (Claude Code, Codex, Cursor, OpenCode, or inside Lemma itself), then verified through the CLI and deployed to a URL with versioned, rollback-able releases. Since v0.8.0 (September 10, 2026) Lemma runs on Windows, and later main-branch development updates merged the teammate workspace into the platform repository, added an in-app sandbox browser with saved logins and browser recordings, made widgets editable files, and brought billing and organization settings into the frontend. Positioned against assembling CrewAI or LangGraph pieces yourself, Lemma trades setup effort for shared state and real governance.
Behind the Verdict
Most agent frameworks give you a graph and a prayer. Lemma gives you an operating boundary. A pod bundles tables and files, serverless functions, agents and workflows, the app people open, and the permissions layer that sits under all of it — and because the agent runs as one shared instance, there is one thing to fix and one set of records everyone reads from. That single design choice is what separates it from a folder of prompts each teammate runs locally. The governance story is unusually concrete for this category. Humans and agents are both pod members subject to the same per-table grants, resource visibility rules, and approval gates. The vendor's own worked example is a refund workflow: a classifier agent may read tickets and draft replies while being blocked from issuing refunds or touching API keys, and closed refunds sit outside its scope. Policy lives in markdown files — searchable, permission-scoped, versioned and read at run time — which makes the rules reviewable instead of buried in prompts. Workflows that hit a human approval step pause and resume later on that person's decision. Where it fits: engineering and ops teams who already drive coding agents and want the output to become a deployed, access-controlled internal tool rather than another local script. Authoring happens in plain files via Claude Code, Codex, Cursor, OpenCode, or inside Lemma, then verification through the CLI and versioned, rollback-able releases behind one URL. Surfaces span Slack and Microsoft Teams, Telegram, WhatsApp, Gmail, a web app, and scheduled runs, with a desktop app for macOS, Linux, and Windows. Where it does not fit: anyone who wants a no-code assistant, a consumer chatbot, or a broad catalog of pre-built enterprise connectors on day one. You need shell fluency. The roadmap has moved fast through 2026 — Windows support only landed in 0.8.0 on September 10, 2026, and the post-0.8.0 main-branch updates (sandbox browser, editable widgets, frontend billing) are explicitly labeled development builds rather than stable releases, so if you are pinning to a version, read the changelog labels carefully. No SLAs or support commitments are stated on the site, which matters if the workflow you are automating is load-bearing.
Researching lemma-platform? Get your full AI stack in 60 seconds.
Free, no signup — tell us your goal and get tools matched to your budget & existing stack.
Real-world workflow fit
Concrete scenarios for the personas lemma-platform actually fits — and what changes day-one when you adopt it.
Stand up a refund pod: connect the support inbox, add a table of open refunds, write a check_policy function that reads refund-policy.md, and gate anything above the threshold on a human approval.
Outcome: Refunds flow through one queue with a logged approver per request, and everyone sees the same status instead of asking in Slack.
Author a pod as plain files locally, verify it through the CLI, then deploy to a versioned URL and connect the Slack surface to the team channel.
Outcome: The agent lives in one place with rollback-able releases, rather than each teammate running their own copy.
Give a teammate a scheduled job that pulls sources, writes findings to a pod table, and posts the weekly digest on a cron schedule.
Outcome: The digest arrives without anyone asking, and the underlying records are there when someone wants to check the source.
Use Cases
- Build a pod that reads incoming email, classifies it, and creates tasks in a table
- Give support teammates and AI assistants one shared refund queue with approval gates
- Ingest files, transform them via deterministic functions, and update a live dashboard
- Run nightly scheduled workflows that generate reports and post them to Slack
- Require a human approval step before an agent issues a refund or grants access
- Let a teammate drive an in-app sandbox browser for sign-in-gated work, with recordings for review
Models Under the Hood
as of 2026-09-22
Limitations
- Open-source runtime (AGPLv3 core, Apache-2.0 SDKs) that you run on your own machine, server, or Lemma Cloud.
- Windows support only landed in 0.8.0 (September 10, 2026), which also carried three backwards-incompatible HTTP API renames ({org_id} → {organization_id}; GET /pods/organization/{organization_id} → GET /organizations/{organization_id}/pods; colon action routes becoming slash routes).
- Direct HTTP API callers must make those three edits — SDK and CLI users only need to upgrade the package.
- The September 16–24, 2026 entries (sandbox browser, editable widgets, teammate workspace, frontend billing) are main-branch development updates, not stable releases, so pin accordingly.
- No SLAs or support commitments are stated on the site.
as of 2026-09-30
Verification history
We have re-verified lemma-platform 8 times since . Each pass re-reads the vendor's own pages and re-checks every listed field against that evidence; passes where nothing had changed are marked as such.
- — re-verified summary, description, our verdict, our analysis, pricing model, pricing tiers, features, integrations, who it suits, who should skip it
- — re-verified summary, description, our verdict, our analysis, pricing model, pricing tiers, features, integrations, who it suits, who should skip it
- — re-verified summary, description, our verdict, our analysis, pricing model, pricing tiers, features, integrations, who it suits, who should skip it
- — re-checked, vendor evidence unchanged
- — re-checked, vendor evidence unchanged
- — re-verified summary, description, our verdict, our analysis, pricing model, pricing tiers, features, integrations, who it suits, who should skip it
Showing the 6 most recent of 8 verification passes.
Free to cite with attribution — this page re-verifies continuously.
12-month cost
Project the real annual outlay, including the implied monthly cost when only an annual tier is published.
Vendor list price only. Add-on usage, seat overages, and contract minimums are surfaced under Hidden costs & gotchas.
Plans compared
For each published lemma-platform tier: who it actually fits, and what it adds vs. the previous tier. Cross-reference the cost calculator above for projected annual outlay.
Open Source Community
$0/mo
Ideal for
Engineering or ops teams comfortable self-hosting and driving a coding agent, who want to build internal tools without a platform subscription.
What this tier adds
Starting tier: the full open-source runtime, self-hosted, with pod architecture, per-table grants, approval gates, Slack surface, desktop app, CLI, and SDKs.
Where the pricing makes sense
The company stage and team size where lemma-platform's pricing actually pencils out — and where peers do it cheaper.
The open-source runtime is free to self-host, so the real cost comparison is your own operational time plus whatever your model provider charges. That puts Lemma in a different bracket from managed no-code agent builders with per-task pricing, and often cheaper than a managed equivalent at the same headcount — the trade is that you supply the setup labor and the hosting.
Setup time & first value
How long it actually takes to get something useful out of lemma-platform — broken out by persona, not the marketing-page minute.
Docs quote about fifteen minutes from a blank pod to an agent that reads real inbound work and writes real records. Realistically, budget 30–60 minutes for a first agent with a connected surface, and an afternoon for a refund-style pod with a policy function, per-table access rules, and an approval gate wired end to end.
Switching to or from lemma-platform
How to bring data in from common predecessors and how to get it back out — written for the switcher, not the buyer.
- →From hand-rolled CrewAI or LangGraph scripts: rebuild the graph as a pod with tables as state, functions as the deterministic nodes, and workflows for the loops.
- →From a local Claude Code or Codex script: import it as the teammate's runtime and keep records in the pod instead of local files.
- →From a shared spreadsheet plus chat bot: move the sheet into a pod table with row-level security and let the agent write to it directly.
- →From a per-person personal AI assistant: consolidate onto one shared running agent so there is one version and one set of records.
- ↗To a managed no-code agent platform: expect to lose the CLI authoring loop and per-table grants, and to rebuild the tools inside a visual builder.
- ↗To a self-hosted framework like LangGraph: re-implement pod tables and permissions yourself, including the approval-pause-and-resume behavior.
- ↗To a general workflow automation tool: rebuild agent steps as fixed automation, which drops the agent's ability to decide what to build next.
Integrations
Resources & Guides
Tutorials & Learning
YouTube returned 6 videos for “lemma-platform”, and we withheld 6: 6 did not mention lemma-platform. We are showing none, because we could not prove any of them are about lemma-platform.
Official links
Featured Head-to-Head Comparisons
Lemma Platform vs Presto Voice
These tools serve entirely different worlds: Lemma is a free, open-source platform for building custom agentic workflows, while Presto Voice is a enterprise voice AI solution for drive-thru automation. Choose Lemma if you need a flexible internal tool for human-AI collaboration; choose Presto Voice if you run a QSR chain and want to automate order taking with proven ROI.
Lemma Platform vs Truleo
Truleo is purpose-built for law enforcement agencies needing to connect siloed data and automate lead generation, report writing, and intelligence briefings. Lemma is an open-source platform for technical teams to build custom agentic workflows in a shared workspace. They serve entirely different markets and use cases, so the right choice depends entirely on your organization: law enforcement vs. engineering/operations.
Lemma Platform vs Locus Robotics
Lemma and Locus Robotics solve entirely different problems—Lemma is for teams building custom AI workflows (open-source, free), while Locus Robotics is for physical warehouse automation (RaaS, enterprise). Choose Lemma if you need a flexible, code-first agentic platform for business processes; choose Locus Robotics if you need proven, scalable mobile robots for high-volume fulfillment.
Popular in Automation & Agents
Air AI
Air's Enterprise Readiness platform turns defense readiness data into a live Readiness Graph instead of slide decks.
Genspark
Genspark is an all-in-one AI workspace where cited AI search turns one prompt into decks, dashboards, docs, and no-code agents.
Cryptohopper
Cryptohopper runs an automated crypto trading bot in the cloud, with Copy Bot mirroring and an MCP server for AI trading agents.
Frequently Asked Questions
Used lemma-platform? Help shape our editorial sentiment research.