Developer

Free
30-day trial for individual developers
  • 50,000 requests / month
  • 5 MCP servers
  • 3 users
  • AI Gateway with failover
  • Basic semantic caching
  • Community support
  • 7-day audit log retention
  • No credit card required
Start Free Trial

Pro Plus

$3,499/mo
For organizations with governance requirements
  • 5,000,000 requests / month
  • 50 MCP servers
  • Up to 100 users (+$15/user/mo)
  • Virtual MCP Servers
  • SSO (Okta, Azure AD, Google)
  • Custom guardrails
  • Full audit logs with export
  • Priority support (4h response)
  • Fine-tuning pipelines
  • 90-day audit log retention
  • Overage: $0.40 / 1K requests
Contact Sales

Enterprise

From $75K/yr
For organizations that need it all, inside their own walls
  • 10,000,000+ requests / month
  • Unlimited MCP servers
  • VPC / air-gapped deployment
  • Multi-region support
  • Self-hosted control plane
  • Custom SLA
  • Dedicated support engineer
  • Compliance-ready (SOC 2, HIPAA, GDPR)
  • Unlimited audit log retention
Contact Sales

* Prices shown are for the WorkingAgents platform. LLM inference costs (OpenAI, Anthropic, etc.) are billed directly by your chosen providers.

Compare Plans

Feature comparison

Feature Developer Pro Pro Plus Enterprise
Price Free (30 days) $499/mo $3,499/mo From $75K/yr
Requests / month 50K 1M 5M 10M+
MCP Servers 5 20 50 Unlimited
Users 3 25 (+$20/user) 100 (+$15/user) Unlimited
AI Gateway
MCP Gateway
Smart Routing
Semantic Caching Basic Advanced Advanced Advanced
CapBAC
Virtual MCP Servers
SSO
Custom Guardrails
Self-Hosted Models
Fine-Tuning
VPC / Air-Gapped
Multi-Region
Audit Log Retention 7 days 30 days 90 days Unlimited
Compliance Readiness
Overage Rate-limited $0.50 / 1K $0.40 / 1K Custom
Support Community Email (24h) Priority (4h) Dedicated engineer
FAQ

Common questions

What counts as a request?

Every API call through the AI Gateway or MCP Gateway counts as one request. This includes model inference calls, tool invocations, and resource reads. Cache hits count as requests but at a reduced rate.

How does the free trial work?

The Developer tier gives you 30 days of full platform access with no credit card required. After 30 days, upgrade to a paid plan to continue using WorkingAgents.

Can I switch plans at any time?

Yes. Upgrade or downgrade at any time. When you upgrade, the new plan takes effect immediately and you're charged pro-rata for the remainder of the billing period. When you downgrade, the change takes effect at the start of the next billing period.

What happens if I exceed my request limit?

On Developer: requests are rate-limited until the trial ends. On Pro: overage is billed at $0.50 per 1,000 additional requests. On Pro Plus: $0.40 per 1,000. On Enterprise: custom arrangements based on your contract.

How does user pricing work?

Each paid plan includes a base number of users. Pro includes 25 users, with additional users at $20/user/month. Pro Plus includes 100 users at $15/user/month for additional seats. Enterprise includes unlimited users.

Is there a free trial for paid plans?

Yes. All paid plans come with a 14-day free trial. Enterprise plans have custom evaluation periods — contact sales to discuss your needs.

How does pricing work for self-hosted models?

Self-hosted model serving is included in Pro and above. You provide the compute (GPUs); WorkingAgents handles deployment, scaling, and optimization. There's no per-token charge for self-hosted models — you only pay the WorkingAgents platform fee and your own infrastructure costs.

What about data egress charges?

WorkingAgents doesn't charge for data egress because your data doesn't leave your infrastructure. Standard cloud provider egress charges may apply for traffic between your services, but that's between you and your cloud provider.

Start building governed AI agents today

30-day free trial. No credit card. Full platform access in minutes.

Start Free Trial Explore the Platform