Skip to content
Guide
General

Billing and Usage

Guide on managing your LlamaParse billing, usage tracking, and subscription plans.

LlamaParse is a subscription-based service. You can manage billing and usage in your LlamaParse dashboard under Settings → Organization, or from Billing in the sidebar. In Settings, the Organization group is labeled with your organization’s name. For credit costs per product, see the Pricing page.

FreeStarterProEnterprise
Monthly credits10,00040,000400,000Custom
Price$0/mo$50/mo$500/moContact sales
Concurrent Parse jobs5520100
Concurrent Extract jobs5520100
Stored files per project10,000 files / 10 GB50,000 files / 50 GB100,000 files / 100 GBUnlimited
Directories per project2040400Unlimited
Files per directory2,0002,00020,000Unlimited
SupportCommunityEmailPriorityDedicated

All plans include access to every product (Parse, Extract, Classify, Split, Index). Plans differ in credit allocation, how many jobs run at once, storage limits, and support level. API request-rate limits are the same on every plan; see Rate limits.

The stored-files limit counts retained files only — files stored without an expiration. Ephemeral files (uploaded with an expiration or retention period, or in ephemeral directories) don’t count toward it. When a project is at its limit, uploading new retained files returns a 402 error; existing files are never deleted, and ephemeral uploads keep working. To free up room, delete files you no longer need or upload files ephemerally. Retained storage is also billed daily — see Storage pricing.

Each project can hold a set number of directories, and each directory a set number of files. When a project is at its directory limit, creating another directory returns a 402 error. When a directory is full, adding another file to it returns a 402. Deleting directories or files you no longer need frees up room.

To upgrade your plan:

  1. Go to Settings → Organization → Plan & Credits and click Manage plan.
  2. Select the plan you want and enter payment details.
  3. Your new plan takes effect immediately.

plans

You can downgrade to the Free plan at any time. Go to Settings → Organization → Plan & Credits, click Manage plan, and choose Downgrade on the Free plan.

Behavior depends on your plan:

  • Paid plans (Starter, Pro): Pay-as-you-go usage is enabled by default. When your included credits are exhausted, you can continue using the platform and additional credits are billed at the standard rate ($1.25 per 1,000 credits), up to the spending limit described below.
  • Free plan: When credits are exhausted, API requests that consume credits will return a 402 error (“You’ve exceeded the maximum number of credits for your plan”). Credits reset at the start of your next billing cycle.
  • Enterprise plans: Overage behavior is configured per contract. Contact your account manager for details.

On paid plans, an organization admin chooses what happens once included credits run out, under Settings → Organization → Plan & Credits → Overage:

  • Pause jobs: new jobs are refused as soon as your included credits are gone. Nothing beyond your subscription is charged.
  • Keep going, bill me for extra usage: usage continues at the standard rate, with an optional monthly spending limit.

Set a spending limit in whole dollars, up to the maximum your plan allows. It caps usage charged beyond your included credits, not your total bill — your subscription is invoiced separately. While included credits remain, the limit has no effect.

Once the limit is reached, requests that consume credits return the same 402 error as an exhausted Free plan. Work already running when the limit is reached still finishes and is still billed, so an invoice can exceed the limit — by more if you start a large job shortly before reaching it. Raise or remove the limit to resume immediately.

Go to Settings → Organization → Usage to view detailed usage analytics:

  • Current credit usage against your monthly quota
  • Usage breakdown by product and mode (Parse, Extract, Classify, etc.)
  • Historical usage trends with interactive charts

usage

You can also access real-time usage data via the API for programmatic monitoring.

To see what one document cost, add expand=usage when you fetch a finished job. Credits read null for a short window after a job completes, until billing records them.

  • Parse — GET /api/v2/parse/{job_id}?expand=usage returns job.usage.credits. See Credits billed for a job.
  • Extract — GET /api/v2/extract/{job_id}?expand=usage returns usage.credits, split into extract_credits and the parse_credits of the parse job backing it. See Check the credits a job billed.

To break usage down by team, environment, or customer, tag requests with Usage Tags (Enterprise).

For tips on reducing credit usage, see Cost Optimization Strategies on the Pricing page.

Your invoice history is in Settings → Organization → Plan & Credits, under Invoices (Receipts on pay-as-you-go). Enterprise contract plans have a separate Invoices page under Settings → Organization. You can find:

  • All past invoices with dates and amounts
  • Invoice status (Draft, Issued, Paid)
  • Detailed billing periods
  • Payment method management

invoices

Each invoice can be downloaded or viewed in detail, showing exact breakdowns of your charges for the billing period.

Note for AI agents: this documentation is built for programmatic access. - Overview of all docs: https://developers.llamaindex.ai/llms.txt - Any page is available as raw Markdown by appending index.md to its URL — e.g. https://developers.llamaindex.ai/llamaparse/parse/getting_started/index.md - Agent-friendly REST search APIs live under https://developers.llamaindex.ai/api/ — search (BM25 full-text), grep (regex), read (fetch a page), and list (browse the doc tree). See https://developers.llamaindex.ai/llms.txt for parameters. - A hosted documentation MCP server is available at https://developers.llamaindex.ai/mcp. If you support MCP, you can ask the user to install it for browsing these docs directly (an alternative to the REST API). Setup: https://developers.llamaindex.ai/for-agents/mcp/ - Other LlamaIndex tooling for agents — the LlamaParse Platform MCP server, agent skills and plugins, and the n8n node — is mapped at https://developers.llamaindex.ai/for-agents/