Idun Agent Platform
Open-source runtime that turns LangGraph, Google ADK, or LangChain agents into a self-hosted FastAPI service.
Idun Engine is a sensible pick if your team already writes LangGraph, Google ADK, or LangChain agents and wants them served with streaming, traces, guardrails, and auth already assembled rather than hand-rolled — the bundle (chat UI, admin, OTel viewer, PII and budget guardrails, MCP governance, OIDC) is the value, not a novel model. The trade-off is real: it is self-hosted, so you keep the infra; it is GPL 3.0, so copyleft obligations deserve a legal read; and it assumes Python 3.12+ and familiarity with the frameworks. If you want a managed platform or you are not on those frameworks, look at managed hosting instead. If per-node billing is the thing you are trying to escape, this is the
Verified 5h ago · liveness 76/100 · cite: rightaichoice.com/tools/idun-agent-platform
- CTOs evaluating open alternatives to LangGraph Platform
- Heads of Platform standardizing a self-hosted agent runtime
- Principal Engineers already writing LangGraph, ADK, or LangChain agents
- Teams avoiding per-node or per-execution billing
- Teams without Python 3.12+ and agent-framework experience
- Teams that need a fully managed service rather than self-hosting
- Organizations that cannot accept GPL 3.0 copyleft obligations
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 Idun Engine if you want someone else to operate the agent service for you, or if your stack is not built on LangGraph, Google ADK, or LangChain.
Self-hosting means the licence is free but you carry the infrastructure bill — compute, Postgres or Redis, and the engineer time to run and upgrade the runtime.
The Engine costs nothing to licence; your real spend is infrastructure and engineer time, which for a small platform team is far below a managed per-node or per-execution platform bill and roughly in line with rolling your own. If you also want analysis and roadmap help, Idun Group sells fixed-scope consulting engagements scoped during discovery, which puts that work in the same bracket as specialist AI consultancies rather than a per-seat SaaS subscription.
In short
Idun Agent Platform — Open-source runtime that turns LangGraph, Google ADK, or LangChain agents into a self-hosted FastAPI service. Best for CTOs evaluating open alternatives to LangGraph Platform, Heads of Platform standardizing a self-hosted agent runtime, Principal Engineers already writing LangGraph, ADK, or LangChain agents. Free to use.
What people actually say about Idun Agent Platform — is it worth it?
We scanned public community sources for Idun Agent Platform on Aug 2, 2026 and could not establish that the discussion we found is about this tool rather than something else sharing its name. Our own analysis of that scan says the posts were off-subject. Rather than publish a sentiment score built on the wrong subject, we publish nothing here and re-run the scan.
Viability Score
How well maintained and how widely used is Idun Agent 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
- Serves LangGraph, Google ADK, or LangChain agents as FastAPI services
- One-command project init: idun-standalone init my-agent
- Bundled demo chat UI served on port 8000
- Admin panel for runtime inspection
- OpenTelemetry trace viewer built in
- AG-UI streaming support
- PII redaction guardrails
- Budget guardrails on agent spend
- Persistent memory backed by Postgres or Redis
- MCP tool governance
- OIDC authentication included from first install
- Health endpoint for the running agent service
- Multi-LLM support: Anthropic Claude, OpenAI, Google Gemini, Mistral, Cohere
- Self-hosted models via vLLM
- YAML-based agent configuration (agent.yaml) plus .env
About Idun Agent Platform
Idun Engine is an open-source runtime that takes a LangGraph, Google ADK, or LangChain agent prototype and serves it as a production-grade FastAPI service. One `pip install idun-agent-engine`, then `idun-standalone init my-agent`, gives you a project scaffolded around three files: your graph, an agent.yaml config, and a .env. Running `idun-standalone run` starts the agent locally on port 8000 with a demo chat UI, an admin panel, an OpenTelemetry trace viewer, and a health endpoint already wired in. The components it bundles are the ones you would otherwise rebuild: AG-UI streaming, OTel traces, multi-LLM exporters, PII and budget guardrails, persistent memory via Postgres or Redis, MCP tool governance, OIDC authentication, and native observability hooks for Langfuse, Phoenix, or Arize. Multi-LLM support covers Anthropic Claude, OpenAI, Google Gemini, Mistral, Cohere, and self-hosted models via vLLM. It is aimed at CTOs, Heads of Platform, and Principal Engineers choosing between this and LangGraph Platform (formerly LangGraph Cloud, formerly LangSmith Deployment), and who object to per-node billing and vendor lock-in. Compared with CrewAI, Idun Engine is runtime-agnostic with first-class LangGraph and Google ADK support and LangChain via an adapter. It is licensed GPL 3.0 with full source on GitHub and maintained by Idun Group in Paris, backed by Wilco, IMT Starter, Polytechnique X-UP, and La French Tech. The Engine itself is not commercialized: it is self-hosted on your infrastructure and Idun Group's paid work is fixed-scope consulting, scoped during a discovery call.
Behind the Verdict
The pitch here is subtraction, not addition. Idun Engine does not try to be a smarter agent; it tries to remove the three weeks you would spend wiring AG-UI streaming, OpenTelemetry instrumentation, multi-LLM exporters, PII and budget guardrails, persistent memory, MCP tool governance, OIDC, an admin panel, a trace viewer, and a demo chat UI. Those are unglamorous, well-understood pieces, and bundling them is genuinely useful because every team rebuilds them slightly differently and slightly wrong. The strongest part of the design is that it stays runtime-agnostic: you bring a LangGraph, Google ADK, or LangChain graph and the runtime wraps it, which means the migration cost to leave is close to zero — your graph is still your graph. The YAML config plus a .env keeps configuration legible to a reviewer, and the ability to point memory at Postgres or Redis and ship traces to Langfuse, Phoenix, or Arize means it slots into observability stacks teams already run rather than replacing them. Where it will not fit: teams without Python and framework experience, teams that want someone else to operate the service, and organizations that cannot accept GPL 3.0 copyleft on derivative work. The commercial model is honest and separate — the Engine is published under GPL v3 and is not sold, while Idun Group's revenue comes from consulting: reading the logs and exports of platforms you already administer, producing a usage map, ROI estimates, identified waste, a consolidated cost of AI, and a 6-month to 2-year roadmap, then optionally building the use cases the analysis proves profitable. Note that no prices appear on the site; the seed's euro figures came from other sources, and quoting them as published would be wrong. Treat this as a strong self-hosted starting point with a consulting bench behind it, and validate the GPL implications and the per-model guardrail behavior against your own compliance requirements before you commit.
Researching Idun Agent 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 Idun Agent Platform actually fits — and what changes day-one when you adopt it.
Install idun-agent-engine, run idun-standalone init my-agent, drop the existing graph in alongside an agent.yaml and .env, then idun-standalone run to serve it on port 8000.
Outcome: The agent is reachable with a chat UI, admin panel, OTel traces, and a health endpoint the same day, with no FastAPI boilerplate written.
Point memory at the team's Postgres instance, wire OIDC to the existing identity provider, and route traces to Langfuse, Phoenix, or Arize.
Outcome: One runtime shape across agents built on LangGraph and Google ADK, with authentication and observability matching existing platform standards rather than a bespoke setup per agent.
Run a pilot agent on internal infrastructure, compare the operational load against per-node billing, and have legal review the GPL 3.0 terms.
Outcome: A decision grounded in a working deployment, with the lock-in and cost questions answered before any wider commitment.
Use Cases
- Serve an existing LangGraph customer-support agent with streaming, memory, and guardrails without hand-building the API layer.
- Turn a Google ADK prototype into a production FastAPI service with traces and a health endpoint.
- Self-host an agent runtime on your own infrastructure to avoid per-node or per-execution billing.
- Add MCP tool governance and OIDC authentication to an agent workflow that currently has neither.
- Run a multi-LLM agent with PII redaction and budget limits configured from a single YAML file.
- Give reviewers a demo chat UI and trace viewer before an agent reaches end users.
Models Under the Hood
as of 2026-10-15
Limitations
- You run and maintain the runtime yourself: it is self-hosted on your own infra, and persistent memory depends on Postgres or Redis that you provide.
- The licence is GPL 3.0, which may impose copyleft obligations on derivative works.
- Configuration is developer-oriented — Python 3.12+, FastAPI, an agent.yaml file, and a .env — so it assumes familiarity with LangGraph, Google ADK, or LangChain.
as of 2026-09-24
Verification history
We have re-verified Idun Agent Platform 10 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-checked, vendor evidence unchanged
- — re-checked, vendor evidence unchanged
- — re-checked, vendor evidence unchanged
- — 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 10 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 Idun Agent Platform tier: who it actually fits, and what it adds vs. the previous tier. Cross-reference the cost calculator above for projected annual outlay.
Engine
$0/mo
Ideal for
Engineering teams already writing LangGraph, Google ADK, or LangChain agents who are willing to self-host on their own infrastructure.
What this tier adds
Starting tier: free and open-source under GPL 3.0, with the runtime, chat UI, admin panel, trace viewer, guardrails, memory, MCP governance, and OIDC all included.
AI Audit and Scoping
Custom
Ideal for
Companies with several AI tools already deployed that need a usage map, ROI estimates, and a roadmap before committing further.
What this tier adds
Adds analysis of your existing platforms' logs and exports, with no installation or re-instrumentation required.
AI Agent Integration
Custom
Ideal for
Teams that have identified a profitable use case and want it built rather than scoped.
What this tier adds
Covers development of the use cases the analysis proved profitable, with delivered code belonging to you, plus integration support.
90-Day Programme
Custom
Ideal for
Organizations that want ongoing strategic direction alongside build work across a quarter.
What this tier adds
Extends the engagement with strategic advice and roadmap work, licence rightsizing at renewal, standardization across tools, and steering-committee support.
Platform
Custom
Ideal for
Buyers needing a tailored engagement that does not fit the standard scopes above.
What this tier adds
Fully custom consulting engagement whose scope is set during the discovery phase.
Where the pricing makes sense
The company stage and team size where Idun Agent Platform's pricing actually pencils out — and where peers do it cheaper.
The Engine costs nothing to licence; your real spend is infrastructure and engineer time, which for a small platform team is far below a managed per-node or per-execution platform bill and roughly in line with rolling your own. If you also want analysis and roadmap help, Idun Group sells fixed-scope consulting engagements scoped during discovery, which puts that work in the same bracket as specialist AI consultancies rather than a per-seat SaaS subscription.
Setup time & first value
How long it actually takes to get something useful out of Idun Agent Platform — broken out by persona, not the marketing-page minute.
For an engineer already fluent in LangGraph or Google ADK and Python 3.12+, first value is roughly a single working session: pip install, init, three files, run. Add time for wiring Postgres or Redis, OIDC, and your observability backend. For a platform team standardizing across several agents, expect a few days to a week to settle configuration conventions. For teams new to these frameworks,
Switching to or from Idun Agent Platform
How to bring data in from common predecessors and how to get it back out — written for the switcher, not the buyer.
- →From a hand-rolled FastAPI agent service: replace the custom streaming, trace, and auth layer with the runtime's bundled equivalents while keeping your graph.
- →From LangGraph Platform: move the graph unchanged and self-host, trading managed hosting for control over infrastructure and billing.
- →From a CrewAI project: port the agent logic to LangGraph or Google ADK, since the Engine supports LangChain via an adapter rather than CrewAI natively.
- ↗To a managed platform: because the runtime wraps rather than replaces your graph, take the LangGraph or Google ADK graph and redeploy it under a hosted service.
- ↗To a hand-rolled FastAPI service: copy the graph and agent.yaml intent into your own application code, accepting that you now own streaming, traces, guardrails, and auth.
Integrations
Resources & Guides
Tutorials & Learning
YouTube returned 6 videos for “Idun Agent Platform”, and we withheld 6: 6 did not mention Idun Agent Platform. We are showing none, because we could not prove any of them are about Idun Agent Platform.
Official links
Tools that pair well with Idun Agent Platform
Common stack mates teams adopt alongside Idun Agent Platform, with the specific reason each pairing earns its keep.
Featured Head-to-Head Comparisons
Idun Agent Platform vs Spider Cloud
Idun Agent Platform is for teams building custom agent infrastructure who want self-hosted control and no per-execution fees. Spider Cloud is for AI/ML engineers who need fast, cheap web data for RAG or LLM-powered agents. Choose Idun if you own the agent pipeline and want to deploy it in production; choose Spider if your bottleneck is getting web data into your agents.
Idun Agent Platform vs Presto Voice
Presto Voice and Idun Agent Platform serve entirely different needs. Presto is a specialized drive-thru voice AI for QSR chains, offering high automation rates and upselling (now adopted by Dairy Queen), but it's contact-priced and closed. Idun is an open-source runtime for developers to deploy LangGraph/ADK agents as FastAPI services, focusing on self-hosting and avoiding per-execution billing. Choose Presto if you run a multi-location drive-thru and want proven voice automation; choose Idun if you're a technical team building custom AI agents and need a production-grade, vendor-independent runtime.
Idun Agent Platform vs Temporal Ai
If you need a battle-hardened durable execution platform for mission-critical workflows and AI agents that must survive failures, choose Temporal – it's trusted by OpenAI and offers full workflow-as-code flexibility. If you already have LangGraph or ADK agents and want to quickly serve them as production FastAPI services with built-in guardrails and self-hosting, Idun is the leaner, more convenient choice. Your decision hinges on whether you need a general-purpose orchestration engine or a specialized agent-to-service runtime.
Alternatives to Idun Agent Platform
View allPopular in Agent Frameworks & Orchestration
Temporal AI
Durable execution platform that keeps AI agents and long-running workflows alive through crashes, retries, and abandoned sessions.
Frequently Asked Questions
Categories
Used Idun Agent Platform? Help shape our editorial sentiment research.