xpander.ai
Also known as: xpander, Xpander
Backend as a service and runtime for AI agents, providing memory, a 2,000 plus tool and MCP repository, multi agent orchestration, and governed deployment.
xpander.ai is a backend as a service and runtime for AI agents, a full lifecycle platform that covers building, deploying, governing, monitoring, and iterating on agents in one product. It positions itself as an agentic native production platform, distinct from workflow automation tools that bolted agent nodes onto trigger and action engines and from build only frameworks that stop at the prototype boundary and leave teams to assemble deployment, monitoring, and governance themselves. The pitch is removing that assembly tax. Founded around 2024 and headquartered in Tel Aviv, xpander.ai gives teams the infrastructure to run reliable, governable agents in production.
You build agents in a Workbench through a chat driven studio with no code, low code, and code first paths, define multi agent collaboration, attach tools, and simulate and debug behavior before deploying. Every agent ships with a harness that manages memory, context compaction, and task state, so agents can work through long running tasks that span hours, pause for human approval, checkpoint their state, and resume without losing context. Its Agent Graph System lets agents branch, parallelize, loop, and coordinate at runtime rather than only chaining linear steps, and a built in sandbox lets agents write and run code at runtime. The company calls the result a task execution engine, not a conversation router.
Agents draw on a repository of more than two thousand MCP connectors and tools, plus custom tools for any private or proprietary API, with Agentic RAG to search within large API responses. They can be invoked from anywhere: API, SDK, MCP, webhooks, Slack, CI/CD pipelines, cron triggers, or other agents. Deployment flexibility is a core selling point, with xpander cloud or self deployment on your own VPC, on premises, or air gapped infrastructure, running on any Kubernetes cluster across AWS, Azure, and GCP. Production operations include semantic versioning, canary and blue green rollouts, automated rollback, and built in governance and observability.
Pricing starts with a free tier that needs no credit card and includes core features, over fifty tools, five serverless agents, 100 interactions and 100 actions a month, token allowances, and five builder seats; agents pause when you hit the caps. Pay as you go removes caps with usage billed monthly in packs, for example around ten dollars per 100,000 agent actions and per 200,000 interactions, with token and resource rates on top. A Team plan adds higher limits and governance for production, and a Custom plan adds full self hosting, SSO, an SLA, onboarding, and tailored resources, with bring your own license available to cut token costs. Billing runs through Stripe or AWS Marketplace.
Vendor details
Canonical URL
https://xpander.ai
Category
Agent infrastructure
Subcategory
Agent runtime and backend
Funding status
Independent. Headquartered in Tel Aviv, Israel, with around 50 employees and roughly $3M raised. Founded around 2024.
Company status
independent
Use cases & customers
Primary use cases
Target customers
Deployment options
Integrations
Agents draw on a repository of more than 2,000 MCP connectors and tools, plus custom tools for any private API, with Agentic RAG to search within large API responses. They are invocable from API, SDK, MCP, webhooks, Slack, CI/CD pipelines, cron triggers, or other agents, with unified event streaming from sources like Slack and Teams.
In practice
Your team prototyped an agent in a framework but stalled for months building deployment, monitoring, and rollback. You rebuild it on xpander, which ships the harness, versioning, and CI/CD so it reaches production fast.
You run a long, multi step task that spans hours and must pause for a human approval midway. The agent harness checkpoints state and resumes without losing context after the approval.
A regulated enterprise cannot send agent data to a vendor cloud. You self deploy xpander air gapped on your own Kubernetes with your own models, keeping execution and data inside your boundary.
Sources & related URLs
Related / legacy domains
Capability coverage
10.5 / 14 capabilities · 75%
| Integrations & Tool CallingCore runtime capability. A repository of more than 2,000 MCP connectors and tools, custom tools for any private or proprietary API, and built in connectors that reach any system, with tool and function calls as the agent's primary actions. | Full |
|---|---|
| Workflow OrchestrationCore product. The Agent Graph System lets agents dynamically branch, parallelize, loop, and coordinate at runtime, with multi agent orchestration, fault tolerance, resumability, and stateful execution including checkpointing and retries, functioning as a task execution engine. | Full |
| Knowledge Grounding & RAGProvides Agentic RAG to search efficiently within large API responses and vector objects with vector database access, a grounding capability, though RAG is framed around tool and API responses more than a turnkey document knowledge base. | Partial |
| Human Oversight & GuardrailsBuilt in human in the loop pause and resume lets agents pause for human approval mid task, and governance controls data, tool, and execution permissions, though there is no dedicated runtime content moderation or guardrail evaluator. | Partial |
| Security, Identity & GovernanceCentralized governance of data, tools, permissions, and execution from one place, SSO on enterprise, an agent registry, and air gapped, on premises, and multi tenant deployment for regulated industries, a strong governance posture, though specific compliance certifications are not detailed. | Partial |
| Observability & AuditabilityBuilt in observability, monitoring, and lifecycle management with external logging integration on enterprise, though observability depth versus dedicated tracing platforms is not fully detailed. | Partial |
| Memory & State PersistenceEvery agent ships with a harness that manages memory, context compaction, and task state, with stateful execution and checkpointing that let agents pause and resume long running work without losing context, a core built in pillar. | Full |
| Deployment & Data ResidencyStrongest deployment flexibility in the category: native on premises, air gapped, and VPC deployment as standalone options, self hosting on your own infrastructure, and running on any Kubernetes cluster across AWS, Azure, and GCP with your own GPUs and models. | Full |
| Prebuilt Agents, Templates & PacksReady to use agent templates for tasks like chat support and code review, a default starter agent, and an agent registry, reusable starting points, though not a deep marketplace of installable production agents. | Partial |
| Triggers & Channel CoverageAgents are invocable from API, SDK, MCP, webhooks, Slack, CI/CD pipelines, cron triggers, and other agents, with unified event streaming and human interfaces across Slack, Telegram, WhatsApp, and Claude, comprehensive trigger and channel coverage. | Full |
| Model Flexibility & RoutingVendor neutral with a built in LLM AI gateway, the ability to bring any AI model, and bring your own license for major providers to reduce token costs, working across multiple AI frameworks. | Full |
| APIs, SDKs & MCP ExtensibilityA REST API, SDK, and CLI, plus deep MCP support including a 2,000 plus connector repository, Cloud MCP servers, and custom tool creation for any private API give comprehensive extensibility. | Full |
| Testing, Debugging & OptimizationVisual testing and simulation of multi agent behavior, debugging of state transitions and agent handoffs, a testing sandbox, CI/CD test pipelines, and A/B testing on enterprise, though it is built in testing rather than a dedicated evaluation product with scorers and judges. | Partial |
| Browser & Computer UseThe generalist agent comes with its own computer and a built in sandbox supports CLI and code execution, letting agents write and run code and operate a compute environment at runtime, though this is code and CLI execution rather than full GUI browser automation. | Partial |
Pricing
Free (no card) · pay as you go (usage in packs) · Team & Custom tiers
usage
Included quota
Free: 100 interactions, 100 actions, 1M input + 100K output tokens, 100 threads, 5 serverless agents, 10 vector objects, 5 builder seats/month. Pay as you go: usage billed in packs (~$10/100K actions, ~$10/200K interactions, $2.5/1M input tokens, $10/1M output tokens, plus thread/vector/agent rates). Team: higher limits + governance. Custom: full self hosting, SSO, SLA, onboarding, A/B testing, external logging, vector DB access.
What is public
The free tier limits and pay as you go per unit rates are public; Team and Custom dollar pricing are not.
Billing mechanics
Free tier with monthly caps that pause agents, then pay as you go billed in fixed monthly packs across multiple usage meters. Team and Custom are higher tiers, the latter sales contracted with full self hosting.
Cost watchouts
Multiple meters bill in parallel (actions, interactions, input and output tokens, threads, vector objects), and always on embedded agents consume dedicated compute continuously; without bring your own license, token spend at scale can dominate the bill.
Variable cost rationale
Beyond the free tier, cost is purely usage based across many meters at once: agent actions, interactions, input and output tokens, threads, vector objects, and running agents. Production agents that call tools heavily, process long contexts, or run always on embedded containers accumulate cost across several axes, though bring your own license can reduce token spend and spend caps limit overages.
Additional watchouts
Usage is metered across many axes simultaneously, so cost can come from actions, interactions, tokens, threads, vectors, and always on embedded agents at once. Always on embedded agents carry dedicated compute cost; model usage at scale benefits from bring your own license.
Overage / add-ons
Free tier caps pause agents at the monthly limit; adding a card activates pay as you go, which removes caps and bills usage monthly in fixed packs. You can set a spend cap that throttles or pauses agents to prevent overages.
Sales call required
No — self-serve available
Free / trial
Free tier (no credit card, indefinite): core features, 50+ tools, 5 serverless agents, 100 interactions and 100 actions/month, 1M input + 100K output tokens, 100 threads, 10 vector objects, 5 builder seats. Agents pause at caps.
Lowest paid plan
Pay as you go: usage billed monthly in packs (e.g. ~$10/100K actions, ~$10/200K interactions, $2.5/1M input tokens, $10/1M output tokens)
Commercial notes
Self serve, developer first with an indefinite free tier and pay as you go, scaling to Team for production governance and Custom for enterprise self hosting and compliance. Billing through Stripe or AWS Marketplace. Vendor neutral with bring your own model and license.
Key ambiguities
Team and Custom pricing and the complete pay as you go pack rate card are not fully public.
Cancellation / refund
Free and pay as you go are self serve. Custom and self hosting are contractual.
Support SLA / resale
Community and self serve on Free and pay as you go; governance on Team; dedicated support with an SLA on Custom. Available via AWS Marketplace.
Missing data
Team and Custom dollar pricing, and the full pack rate card for threads, vector objects, and agent containers, are not public.
Related vendors
- AgentOps — Agent observability and reliability platform with broad model and…
- Agno — High-performance agent runtime and framework (formerly Phidata) with…
- Apify — Cloud platform for web scraping and automation with 45,000+ prebuilt…
- Arcade — Authenticated tool calling platform and MCP runtime that handles…
- Arize AI — AI observability and evaluation platform that traces, evaluates, and…
- Braintrust — AI evaluation and observability platform with self-serve pricing,…