A low-level orchestration framework for building stateful, multi-step agent workflows with explicit control over state, branching, and human-in-the-loop steps.
Agent Development comparison
Pricing, pros, cons, and ideal use cases — side by side.
A low-level orchestration framework for building stateful, multi-step agent workflows with explicit control over state, branching, and human-in-the-loop steps.
Microsoft's open-source agent framework for Python and .NET, the 1.0 successor to both Semantic Kernel and AutoGen, with graph-based workflows and hosting on Foundry Agent Service.
| LangGraph | Microsoft Agent Framework | |
|---|---|---|
| Pricing | FreemiumLangGraph itself is open source under the MIT licence and free to use. Managed hosting is LangSmith Deployment: the LangSmith Plus plan is $39 per seat per month (billed monthly), includes 10,000 base traces a month (then 0.005 LSU per base trace) and one free Small serverless deployment; additional Serverless and Dedicated deployments are metered in LSUs ($1.00 each) at 0.0675 LSU per runtime vCPU-hour, 0.0090 LSU per runtime GiB-hour, 0.177 LSU per database vCPU-hour and 0.025 LSU per database GiB-hour, so a Small serverless deployment (0.8 vCPU, 2 GiB) costs 0.072 LSU per running hour. Self-hosted and hybrid deployment are Enterprise options at custom pricing. As read on the vendor's pricing page on 10 November 2026. | FreeThe framework is MIT-licensed and free. Hosting is optional and separate: on Foundry Agent Service there is "no additional charge for creating or running Foundry-native agents using prompts and workflows", while hosted agents bill container compute at $0.0994 per vCPU-hour and $0.0118 per GiB-hour (West US 3), plus model tokens and tool fees such as Web Search at $14 per 1,000 transactions. As read on the vendor's pricing page on 11 November 2026. |
| Category | Agent Development | Agent Development |
| Ideal for | Engineering teams building multi-step agentsEnterprise platform teamsWorkflows needing human-in-the-loop control | .NET and Python teams building agents on AzureTeams migrating off AutoGen or Semantic KernelEnterprises that want explicit, graph-based multi-agent workflowsMicrosoft-ecosystem shops hosting agents on Foundry |
Microsoft Agent Framework is the lighter-weight option (Free), while LangGraph sits higher on the pricing ladder (Freemium). LangGraph is built around engineering teams building multi-step agents; Microsoft Agent Framework leans more toward .net and python teams building agents on azure. Shortlist the one whose strengths line up with your biggest constraint.
Get one AI workflow a week showing Agent Development in a real stack — what they cost, and where each one breaks.