Cost Estimates
Foundry’s infrastructure cost scales with usage. The platform costs are modest at evaluation and early production tiers; AI API usage is the primary variable cost.
Infrastructure costs
Section titled “Infrastructure costs”| Service | Plan | Monthly Estimate | Notes |
|---|---|---|---|
| Vercel | Pro | ~$20 | Compute offloaded to Convex and Cloudflare. Vercel handles SSR, Edge Middleware, and OAuth callback routes. |
| Cloudflare Workers | Paid ($5/mo) | ~$5-15 | Agent worker is low-volume (structured analysis requests). Sandbox worker costs depend on container usage. |
| Convex Cloud | Free / Pro | $0-25 | Free tier covers evaluation. Pro tier ($25/mo) for production with higher function call limits and bandwidth. |
| Clerk | Free / Pro | $0-25 | Free tier covers up to 10,000 MAU. Pro tier at $0.02/MAU beyond that. |
| Total infrastructure | ~$25-65/mo | Evaluation can run on free tiers for all services. |
AI API costs (Anthropic)
Section titled “AI API costs (Anthropic)”AI costs depend on which features you use and how frequently. Here are estimates per operation based on the context assembly pipeline and model tiers.
Per-operation estimates
Section titled “Per-operation estimates”| Operation | Model | Input Tokens | Output Tokens | Est. Cost |
|---|---|---|---|---|
| Document analysis | Opus 4.6 | ~8,000 | ~4,000 | ~$0.30 |
| Task decomposition | Sonnet 4.5 v2 | ~5,000 | ~3,000 | ~$0.04 |
| Subtask generation | Sonnet 4.5 v2 | ~4,000 | ~2,000 | ~$0.03 |
| Sprint planning | Sonnet 4.5 v2 | ~6,000 | ~3,000 | ~$0.04 |
| Gate evaluation | Sonnet 4.5 v2 | ~5,000 | ~2,000 | ~$0.03 |
| Risk assessment | Sonnet 4.5 v2 | ~4,000 | ~2,000 | ~$0.03 |
| Health scoring (per workstream) | Sonnet 4.5 v2 | ~3,000 | ~1,000 | ~$0.02 |
| Daily digest | Sonnet 4.5 v2 | ~2,000 | ~500 | ~$0.01 |
| Skill execution (sandbox) | Sonnet 4.5 | ~10,000 | ~5,000 | ~$0.05 |
Monthly estimates by usage tier
Section titled “Monthly estimates by usage tier”1 program, 10-20 requirements, light usage
| Category | Monthly Estimate |
|---|---|
| Document analysis (5 docs) | ~$1.50 |
| Task decomposition (20 reqs) | ~$0.80 |
| Subtask generation (20 tasks) | ~$0.60 |
| Sandbox executions (10 runs) | ~$0.50 |
| Health scoring + digests | ~$1.00 |
| AI total | ~$4-5 |
| Infrastructure | $0 (free tiers) |
| Grand total | ~$5/mo |
3-5 programs, 50-100 requirements, regular usage
| Category | Monthly Estimate |
|---|---|
| Document analysis (20 docs) | ~$6 |
| Task decomposition (80 reqs) | ~$3 |
| Subtask generation (60 tasks) | ~$2 |
| Sandbox executions (40 runs) | ~$2 |
| Health scoring + digests | ~$5 |
| Sprint planning + gates | ~$3 |
| AI total | ~$20-25 |
| Infrastructure | ~$50 |
| Grand total | ~$70-75/mo |
10+ programs, 200+ requirements, heavy sandbox usage
| Category | Monthly Estimate |
|---|---|
| Document analysis (50+ docs) | ~$15 |
| Task decomposition (200+ reqs) | ~$8 |
| Subtask generation (150+ tasks) | ~$5 |
| Sandbox executions (100+ runs) | ~$5 |
| Health scoring + digests | ~$10 |
| All other AI features | ~$10 |
| AI total | ~$50-60 |
| Infrastructure | ~$65 |
| Grand total | ~$115-125/mo |
At this tier, Convex Pro and Clerk Pro are required. Cloudflare Container costs increase with sandbox concurrency.
Cloudflare container costs
Section titled “Cloudflare container costs”The sandbox worker provisions Docker containers on Cloudflare. Container costs are separate from the Workers Paid plan.
- Compute: Billed per GB-second of container runtime
- Concurrency: Up to 20 concurrent containers per sandbox worker
- TTL: Containers expire after 5-60 minutes (configurable per session)
Short-lived sandbox sessions (5-15 minute TTL) keep container costs low. Long-running interactive sessions (30-60 minute TTL) cost more but are less frequent.
Cost optimization strategies
Section titled “Cost optimization strategies”- Use prompt caching aggressively. Batch sequential AI operations within the same program to maximize cache hits on shared context.
- Set appropriate sandbox TTLs. Default to 15 minutes for automated execution. Reserve 30-60 minute TTLs for interactive debugging sessions.
- Monitor AI usage records. The
aiUsageRecordstable tracks every API call with model, token count, and cost. Use the Agent Activity dashboard to identify expensive patterns. - Leverage free tiers for evaluation. Convex Free, Clerk Free (10K MAU), Vercel Hobby, and Cloudflare Workers Free cover evaluation workloads.
- Right-size model tiers. Opus 4.6 is reserved for document analysis where extraction quality matters. All structured analysis routes use the more cost-effective Sonnet tiers.