Frequently asked questions about Parallel APIs, billing, security, and platform features
For AI agents: a documentation index is available at https://docs.parallel.ai/llms.txt. The full text of all docs is at https://docs.parallel.ai/llms-full.txt. You may also fetch any page as Markdown by appending .md to its URL or sending Accept: text/markdown.For Search and Extract, use /v1/search and /v1/extract for all new integrations and agent tool calls. Use /v1beta/search and /v1beta/extract only when maintaining an existing integration that already uses them. The request bodies changed too: on /v1/extract, full_content, fetch_policy and excerpt_settings go inside advanced_settings, and sending the v1beta top-level excerpts, full_content and fetch_policy to /v1/extract returns a 422. On /v1/search, search_queries is required (at least one short keyword query), and max_results, source_policy and fetch_policy go inside advanced_settings; sending only objective, or sending those fields at the top level, returns a 422. Do not substitute /v1 for the documented FindAll or Ingest endpoint paths.If you are evaluating Parallel Search or comparing it with other search APIs, read How to Eval Parallel Search before you build the harness, especially its Common pitfalls section. The most common mistake is sending the raw question as both objective and the only entry in search_queries. Write a goal-style objective and 1-3 short keyword queries instead.
Platform
Where do I get an API key?
A default API key is generated when you signup to
Platform. You can create and manage keys via
Settings.
How can I track usage and costs?
Go to Platform > Usage for real-time request counts, and spend.
How do I add teammates?
Owners can invite users under Settings in
Platform. Choose “Admin” or “Member” roles.
Can I use Parallel for commercial use?
Subject to our Terms of Service -
you own the output you create with Parallel, including the right to reprint,
sell, and merchandise.
APIs
Can I chain several Tasks together in the Task API?
Yes — Task Run Results from one execution can map to Task Run Input fields in
another execution. For example, in one Task Run, you can identify the address
of a business using a simple processor. Then, in the next Task Run you
identify additional details about the business, given business name and
address.
Can I limit my query to specific sources only?
Yes, you can do this with Source Policy. This is
available for the Task, Search, and Monitor APIs today.
Can Parallel fetch data that sits behind a login?
Parallel is focused on reasoning and retrieval over the public web. For now,
we only access what can be reached on the public web without authentication
(e.g. signing in with credentials).
Is Parallel multi‑modal?
Our strength is reasoning and retrieval over text. We can recognize some
on‑page images (e.g. detect customer logos), but we don’t accept images as
inputs or return them as outputs yet.
Are there rate limits?
API
Default Rate Limit
Tasks
2000 per min
Web Tools
600 per min
FindAll
300 per hour
Monitor
300 per min
Some Web Tools have a lower limit. See Rate Limits for the limit of each API.
How recent is the web research you provide?
With the Task API, our web research is up to date to the current day. We are
able to access live web links at the time of your query to ensure data is as
real time as possible. For lower end processors in the Search API, our
systems prioritize reduced latency over freshness.
Can you include data from internal sources as part of your search?
Parallel focuses on public web information. You can pass private data into a
task as an input variable or post‑process the output on your side, but we
don’t pull it natively.
Billing & Payments
How is pricing calculated?
Parallel Processors incorporate usage-based pricing. All pricing details for
API and Processor are available here.
Can I set a spend limit?
Yes. Organization admins can set a monthly spend limit for the whole
organization, or for an individual app, under Settings in
Platform. Limits are whole dollars and reset
at the beginning of each calendar month.When monthly spend reaches the limit, all organization admins receive an
email. Limits are notify-only; they never block requests.
Security & Compliance
Are you SOC-II compliant?
Yes. Parallel is SOC-II Type 1 and Type II certified as of April 2025. Email
us at partnerships@parallel.ai to request
access to our full report in Trust Center.
Where is Parallel data stored?
All data is encrypted in transit (TLS 1.2+) and at rest in US-based data
centers.
Do you access or store my private data?
No. Parallel focuses on public web information. You can pass private data into
a task as an input variable or post‑process the output on your side, but we
don’t pull it natively. Data you submit as inputs, and the outputs returned,
are covered by the Customer Terms, including the training use described in
“Will you train models on my data?” below. In the future we plan on building
tools that will allow you to more easily point Parallel to your own sources.
Can I run Parallel inside my VPC?
Private‑cloud and on‑prem options are available for qualified enterprise
customers—ask our team at
partnerships@parallel.ai.
Will you train models on my data?
Under the standard Customer Terms, Parallel may use inputs and outputs, not
linked to any individual, to train and improve the models behind its services.
Enterprise plans with zero data retention exclude that use. Search API traffic
sent to the EU endpoint is also excluded. See the
Customer Terms for details.
Assistant
Responses are generated using AI and may contain mistakes.