> ## Documentation Index
> Fetch the complete documentation index at: https://docs.hud.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Limits

> Rate limits, runtime caps, and size limits on the HUD platform, and what happens when you reach one.

The limits you are most likely to reach. Plan-dependent limits are listed for **Pay as you go**,
**Pro**, and **Enterprise**; see [plans](/platform/billing#plans). For limits you set yourself, see
[spend limits](/platform/billing#spend-limits).

## API rate limits

Requests to the [REST API](/platform/rest-api), including those the CLI and SDK make, are limited to
1,000 per minute per member, shared by all of their API keys, with bursts of up to 500. A refused
request returns `429` with a `Retry-After` header; wait that long and retry. The web app and the
[Default key](/platform/credentials#the-default-key) that hosted runs carry are not limited.

When the service is busy, a request can also return `503` with `Retry-After`. Retry it the same way.

## Model calls

Model calls through HUD's gateway are limited per team:

| Plan | Calls per minute |
| - | - |
| Pay as you go | 60 |
| Pro | 300 |
| Enterprise | No limit |

Only a hosted run's calls to the model it is evaluating are exempt. Everything else counts: scripts,
agents that run on your own machine, even against a hosted environment, and calls a hosted run makes
to other models, such as an LLM grader's. Calls with
[your own provider key](/platform/credentials#bring-your-own-provider-keys) count too. A refused
call returns `429`.

## Runs

| Limit | Value |
| - | - |
| Runtime of one environment | 4 hours on Pay as you go and Pro, 24 hours on Enterprise. A task's [`RuntimeConfig`](/v6/reference/runtime#runtimeconfig) can set a shorter `run_timeout_s`. |
| Environment sizes | Large and Extra Large must be enabled for your team. See [EC2 profiles](/platform/environments#ec2-profiles). |

Plans don't set a fixed number of runs a team can have at once.

## Traces

A [trace](/platform/traces) stores the steps a run records and the screenshots, files, and video
attached to them:

| Limit | Value |
| - | - |
| Everything one trace records | 500 MB |
| One step, including what is attached to it | 25 MB |
| Screenshots and other images per trace | 500 |
| Other files per trace, including segments of robot camera video | 1,000 |

Data past a limit is refused with `413` and is missing from the trace. A computer-use agent that
takes one screenshot per step reaches the image limit after 500 steps.

## Files and imports

| Limit | Value |
| - | - |
| One uploaded file | 250 MiB |
| File names | 255 characters |
| Uploads in progress | 5 per member, including unfinished ones. The web app queues the rest. |
| Files deleted at once | 500 |
| Storage | Depends on the plan. See [storage limits](/platform/data#upload-files). |
| Harbor import | A zip of at most 1 GiB, 200,000 entries, and 8 GiB unpacked. See [Harbor import limits](/platform/harbor-import#limits-and-permissions). |
| Imported job runs | 1,000 per call. See [importing jobs](/platform/rest-api#jobs). |


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.