> For the complete documentation index, see [llms.txt](https://lamatic.ai/llms.txt).

# Limits & Quotas

Practical limits for building and running workflows on Lamatic. Use this when designing flows, sizing jobs, or explaining plan differences.

**Enforced** means a hard limit. Exceeding it fails or is blocked in Studio.

  **Recommended** means it isn't always blocked, but ignoring it makes runs more likely to time out, run out of memory, or fail under load.

---

## How long a flow can run

| Mode | Limit | Notes |
|------|-------|-------|
| **Sync** (waits for a response) | ~2 minutes on Pro, ~5 minutes on Enterprise | Best for APIs and widgets that return a result immediately. Keep the flow small. |
| **Async** | Up to ~20 minutes on the standard path | Trigger response type must be set to **async**. Returns a request ID right away, and the flow continues in the background. |
| **Container mode** (opt-in, for memory-heavy flows) | Up to 1 hour | For flows that need more memory. Turn it on at the project or flow level when standard runs are running out of memory. |

**Recommended flow size**

| Mode | Guidance |
|------|----------|
| Sync | ~20 nodes max |
| Async | ~50 nodes max |

Larger flows are harder to debug and more likely to hit time or memory limits. Split big workflows into smaller flows and call them with a **Flow** node.

---

## Memory

| Runtime | Limit |
|---------|-------|
| Standard execution | 128 MB |
| Container mode (opt-in) | 6 GB |

If a flow consistently fails with memory errors, enable **container mode** for that project or flow, reduce batch concurrency, or shrink payloads.

---

## Loops, batches, and agents

| Limit | Value | Type |
|-------|-------|------|
| Nodes inside a loop, batch, or agent body | 20 | Enforced for Starter and Pro. Unlimited on Enterprise. |
| Nodes inside a batch body | 5 or fewer recommended | More nodes per iteration raise memory use. Prefer a sub-flow for larger logic. |
| Batch concurrency (parallel items) | 1 to 50, default 10 | Configurable on the Batch node. |
| Nested Batch nodes | Not allowed | Enforced. |
| Batch iterations (Starter / free) | 100 max | Enforced. |

If you need more than 20 nodes inside a loop or batch, put the extra logic in a separate flow and call it with a **Flow** node.

---

## Code node

| Limit | Value |
|-------|-------|
| Processing time | ~2 minutes on Pro, ~5 minutes on Enterprise |
| Code size | 10,000 characters max |
| Restricted APIs | Node process APIs such as `require`, `eval`, and `child_process` are blocked |

---

## Jobs (data syncs)

Jobs are scheduled syncs from sources such as databases, sheets, and storage.

| Limit | Pro / Starter | Enterprise |
|-------|----------------|------------|
| Active jobs | 10 or fewer recommended | Higher, contact Lamatic |
| Sync capacity | Shared infrastructure (~2 GB) | Dedicated infrastructure with higher capacity |

Need more jobs or larger syncs? Use **Enterprise**.

**Default job processing settings** (Settings → Jobs)

| Setting | Default |
|---------|---------|
| Batch size | 5 |
| Max concurrency | 5 |
| Max retries | 3 |

---

## Plan limits

### Starter

| Resource | Limit |
|----------|-------|
| Flows per project | 5 |
| Starter projects per organization | 1 active |
| Workflow executions | 3,000 per 30 days, 100 per day |
| Vector records / writes | 1,000 each |
| Reports & charts history | 7 days |
| Team members | 3 (unless the org has a Pro project or is Enterprise) |
| AI Assistant | Not available |

### Pro / Trial

| Resource | Limit |
|----------|-------|
| Workflow executions | 100,000 |
| Vector records / writes | 10,000 each |
| Reports & charts history | 30 days |
| Sync flow duration | ~2 minutes of processing time per run |
| AI Assistant | Available |

### Enterprise

| Resource | Limit |
|----------|-------|
| Workflow executions | Unlimited |
| Vector / request caps | Unlimited |
| Sync flow duration | ~5 minutes of processing time per run |
| Nodes inside loops / batches | Unlimited |
| Jobs / sync capacity | Dedicated, higher capacity |
| Reports & charts history | Up to 90 days |

---

## Other limits

| Limit | Value |
|-------|-------|
| Nested Flow calls (flow calling flow) | 100 levels max |
| Large async payloads | Inputs above ~125 KB are stored automatically and still work |
| Flow retries (configurable) | Max retries 5 or fewer, retry delay 30 seconds or less |

---

## Quick tips

1. Keep 5 or fewer nodes inside a Batch. Use a sub-flow for more.
2. Keep sync flows to about 20 nodes, or switch the trigger to async.
3. Keep Code node work under the plan's processing time budget (~2 minutes on Pro).
4. Keep active jobs to about 10 or fewer on Pro. Use Enterprise for more.
5. For memory-heavy flows, enable container mode or reduce batch concurrency and payload size.

---

Hitting a `429` or a failed run? See the [Error Reference](/docs/error-reference) page for what each status code means and how to fix it.