Fully Managed Qdrant
as a Service

Deploy Qdrant as a fully managed service starting at €9/mo. Get automated backups, SSL, updates, support and monitoring included.

Qdrant is an open-source vector database written in Rust. It stores embeddings, filters them by payload, and returns nearest neighbours in milliseconds, which is the retrieval layer behind semantic search, recommendations and RAG assistants.

✓Free 7-day trial  ✓99.9% Uptime SLA  ✓No credit card  ✓Cancel anytime

✓Free 7-day trial  ✓99.9% Uptime SLA
✓No credit card  ✓Cancel anytime

Start free trial
Pricing
Qdrant

Qdrant

STARTING AT

€9/month
Automated Backups
Monitoring
Automated Updates
Auto SSL

USAGE

Unlimited
Human Support
Custom Domains
Terminal Access
File Manager Access
Deploy in your region — 21 locations worldwide
GermanyFinlandNetherlandsUKSwedenUnited StatesCanadaSingaporeJapanAustraliaBrazilSouth Africa+9 more →
Qdrant Preview Image

ABOUT THE SOFTWARE

What is Qdrant

Qdrant is an open-source vector similarity search engine and database. It holds high-dimensional embeddings alongside arbitrary JSON payloads, indexes them for approximate nearest-neighbour retrieval, and answers filtered queries over REST or gRPC in milliseconds.

The engine is written in Rust and released under the Apache License 2.0 by a company based in Berlin. Because the licence is permissive rather than copyleft, and carries no network or managed-service clause, running it as a hosted service is unrestricted. The whole engine is in the open build: sharding and replication, JWT-based role separation, scalar, binary and product quantization, snapshots, on-disk storage and the web dashboard. Only the vendor's own cloud control plane sits outside it. Qdrant is developed and maintained by a company based in Berlin. The project has more than 34,000 GitHub stars and more than 2,700 forks.

Retrieval uses an HNSW graph index, shipped with a graph degree of 16 and a construction beam of 100. Payloads are stored on disk by default so memory tracks vector count rather than document size, and payload indexes make filters part of the graph traversal instead of a post-filter. A single collection can hold dense vectors, sparse vectors and multivector late-interaction embeddings at once. Official client libraries cover Python, JavaScript, Rust, Go, Java and .NET.

FEATURES

What Qdrant does

Qdrant covers the whole retrieval path: index, filter, rank and store. The features below are all present in the open build, so nothing on this page depends on a proprietary tier or a licence key you would have to buy separately.

HNSW vector index

Graph-based approximate nearest-neighbour search with tunable graph degree and construction beam, hardware-accelerated distance computation, and optional GPU-assisted index building.

Payload filtering

Attach arbitrary JSON to every point and filter on keyword, full-text, numeric range and geo conditions. Payload indexes fold the filter into graph traversal rather than discarding results afterwards.

Hybrid and sparse search

Dense, sparse and multivector late-interaction embeddings live in one collection and one query, merged with reciprocal rank fusion or distribution-based score fusion.

Quantization

Scalar quantization cuts vector memory roughly fourfold and binary roughly thirty-twofold, with originals kept on disk. You trade recall for memory per collection, not globally.

Multitenancy

One collection partitioned by a tenant field with a payload index, rather than a collection per customer. JWT tokens scope a key to a single tenant's data.

Snapshots

Per-collection and per-node archives for backup, migration and cloning, writable to object storage. A snapshot restores onto any other Qdrant install without conversion.

Sharding and replication

Distributed mode spreads a collection across nodes with Raft consensus. A replication factor above one allows rolling upgrades without taking search offline.

REST, gRPC and dashboard

An OpenAPI REST interface for debugging and a gRPC interface for batch throughput, plus a built-in web dashboard for browsing collections and running queries.

Software Source
Documentation
Video Tutorials

WHAT'S ALWAYS INCLUDED

Every app. Fully managed.
Nothing extra to pay for.

Every app on DANIAN ships with the same operational baseline. You do not choose a backup add-on or a monitoring tier. Hosting, patching, backups, SSL, and support are part of the €9, and they apply to Qdrant the same way they apply to everything else in the catalog.

Automatic updates and patches

Apps run the latest stable version. Security patches applied silently, with rollback if needed.

Daily off-site backups

Multiple daily backups in redundant off-site locations. One-click restore if anything goes wrong.

24/7 uptime monitoring

Continuous monitoring with instant alerting. We respond before you notice.

SSL, firewall, DDoS protection

Auto-renewing SSL, hardened firewall rules, DDoS mitigation on every deployment.

Performance and scaling

We monitor resource usage continuously. When your app needs more headroom, we flag it and upgrade with your explicit approval.

Dedicated engineering support

Real engineers by email and ticket. DNS, SMTP & migration help. All included in €9.

Start 7-day free trial

WHY MANAGED

Why teams pick managed Qdrant

The hosted vector-database vendors have moved their entry floors up. The cheapest production tiers now start near $45 to $50 a month in platform minimums before a single query is billed, and the meter runs on stored vectors and read units rather than on anything you control.

Running the engine yourself removes the meter and adds an operations job. A default Qdrant build binds to every network interface with no API key configured, which is why exposed instances keep turning up in public scans. Backups are your problem, upgrades must step through one minor release at a time because the storage format is only guaranteed compatible between neighbours, and the process will be killed by the kernel if memory planning is wrong.

The memory planning is where most self-hosted instances fail. Qdrant sizes an in-memory collection as vectors multiplied by dimensions multiplied by four bytes, then roughly 1.5 times that for index and metadata overhead. One million 768-dimension vectors lands near 4.6 GB. The trap is that this figure describes steady state, not ingest: uploading in bulk with quantization enabled can cache full vectors while segments optimise, and the instance dies part-way through a load that the arithmetic said would fit.

We size the instance before the first upload, set the API key and the read-only key on first boot, keep the API and dashboard ports off the public internet, and disable telemetry. Bulk loads run with the indexing threshold set to zero and indexing restored afterwards, so the optimiser is not competing with the upload. Payload stays on disk. Quantization stays off until your own recall testing says it should be on.

REVIEWS

Hear from customers ​like you​​​​​​​

Successful businesses and professionals around the world rely on DANIAN every day
5 stars review

No more maintenance headaches!

Finally, I can relax and not bother with maintenance. Each of the softwares that we run on DANIAN is automated and running stable for couple months already. They even assist us with modifying the DNS records. Once we needed to modify the config of Nextcloud so that we are able to upload higher than 5GB files and they assisted us fast. I also love their documentation!

Review from Kevin Blackwell

Kevin Blackwell

Review Origin United States
5 stars review

Best managed service

I previously ran Peertube, Mastodon, Mattermost and Mirotalk on other providers before switching to DANIAN. I was tired of handling the updates and backups myself. Not only that the performance is "out of this world" fast, but their tech support is amazing. Since then I added 5-10 more Apps on my account. I also love how they add more softwares on their list constantly!

Review from William Northfield

William Northfield

Review Origin United Kingdom
5 stars review

Love the personalized service!

I own an ecommerce startup. I knew that I need some softwares so that I add automations in my biz. I contacted their support team and I explained my needs. Albert recommended me couple softwares, each for a specific use-case. I signed up for the free trial and tested all. Ended up activating 6 apps. Everything is running smooth 2 months so far. I love the personalized service that I get with them!

Review from George Adams

George Adams

Review Origin United States
5 stars review

n8n runs excellent

We needed to add the ffmpeg functionality to our n8n instance. I contacted their support team. They made the required changes in couple minutes and notified me. Usually for this type of service I would need to hire a developer on a hourly basis. Love them!

Review from David Hartmann

David Hartmann

Review Origin Germany
5 stars review

Smooth experience

I needed a way to manage the projects and files for my company. This was the first time I learn about opensource. I asked their support team and they assisted me immediately. I started with Openproject. Later I added Nextcloud. My team were able to get used to the apps fast.

Review from Levi Nilsso

Levi Nilsson

Review Origin Norway
5 stars review

Bye GA

I wanted to replace GA with something but I didn't know the options. I talked with Jack from support. Couple minutes after that, I have a running Matomo Analytics. They even configured the tracking and connected it with my website. I have only words of appreciation and respect!

Review from Wyatt Mitchell

Wyatt Mitchell

Review Origin Canada
5 stars review

Love the freedom

I wanted to use a crm for my startup. I signed up for the free trial. Tried out espocrm and dolibar and picked one. Their dashboard is very easy for a beginner. I recommend them and feel confident in choosing them.

Review from Elias Koch

Elias Koch

Review Origin Germany
5 stars review

Wow just wow

After discovering DANIAN, I wished I'd found them sooner — their solution is excellent and deserves wider recognition. Very good solution!

Review from Tyler Hollander

Tyler Hollander

Review Origin Canada
5 stars review

Professional tech support!

I thought that I knew what I was doing and I made changes in the Mattermost backend config file. Then it broke. I reached out to their tech and received an incredibly specific and fast response. They found the error fast and provided a clear solution.

Review from Bastian Engel

Bastian Engel

Review Origin Germany
5 stars review

Their support team is the best!

I had a problem with my card and the subscription could not be renewed in time. I notified their support and surprisingly, they showed much understanding. They basically left my apps run until I solved the card issue. Other providers would have deleted my data in a heartbeat. This is the way to build client-vendor trust!

Review from George Adams

George Adams

Review Origin United States
5 stars review

Fair price and fast

4 months ago I activated n8n, Uptime kuma, Vaultwarden and Paperlessngx. So far everything is perfect. The performance of all apps is excellent. I contacted their support team couple times for using custom subdomains for the apps and they were fast and kind. I'm constantly recommending them to my network!

Review from Atanas

Atanas

Review Origin Macedonia
5 stars review

Vaultwarden 10/10

I wanted to selfhost a passwords manager for my company but I didn't trust the mainstream providers. I tried out Vaultwarden for the free trial. I loved how their support team communicated with me professionally but still kind. I know that I can contact them whenever with whatever question I have. Recommended!

Review from James Smith

James Smith

Review Origin United States
5 stars review

Multiple locations options

I run an agency. The majority of our clients are in different countries. I love that DANIAN has 20+ countries and I can spin up any app in any location. This gives us the freedom to focus on our clients. Sure, I can employee a dev in my team, but honestly, I prefer to not have those headaches. With DANIAN everything is running 24/7.

Review from Edward Nelson

Edward Nelson

Review Origin United States
5 stars review

Everything just works!

My focus is on my business, not running servers. I love that I can lauch any app without having to think about the tech behind it. Everything just works!

Review from Tony Roy

Tony Roy

Review Origin United States
5 stars review

Awesome support team!

I signed up for the free trial in order to test Chatwoot. I needed to put a livechat widget on my website but I don't have much tech experience. I contacted their support to see if they can assist. They went above and beyond by generating the livechat widget code for me and inserting it in my website. They answered all my additional questions. 

Review from Filip

Filip

Review Origin Austria
5 stars review

Very proactive!

Today couple of the videos on my Peertube got viral. Before I even noticed, their support have notified me via email that they proactively upgraded my app with more cpu & ram so that it continues to work without any downtime. They left the upgraded app with the added resources for couple days until the traffic stabilizes. After that I had the possiblity to stay on the same plan or keep the upgraded plan. Out of gratitude I stayed on the upgraded one.

Review from Luke Hall

Luke Hall

Review Origin Australia
5 stars review

Win-win!

I run a web design biz. After I perform the work for my clients, they constantly ask me for more software solutions. So I partnered with DANIAN. Not only I provide the requested softwares for my clients and get the increased revenue boost by delegating all to DANIAN, but also for those clients that wish to order directly from DANIAN, I send them an affiliate link. Win-win on all sides!

Review from Luke Walker

Luke Walker

Review Origin United Kingdom
5 stars review

Flexible

For the past couple months I was running many apps with DANIAN for my clients. But one of my clients wanted a software that isn't on the current list. I talked with their tech support. They accomodated me and added the software the same day. The next day, I signed for their Partners program. We will go long way together!

Review from Jeffrey Johnson

Jeffrey Johnson

Review Origin United States
5 stars review

Awesome customer service

I needed to add analytics on my website but I didn't know anything about DNS. Their team guided me at all phases and they performed the changes for me. Additionally they were transparent and notified me of every change applied and steps performed.

Review from Gabriele Marino

Gabriele Marino

Review Origin Italy
5 stars review

Good admin dashboard

Not only that I got Chatwoot, I also have access to the management dashboard. I love that I can view my hourly/daily usage and perform manual backups and restorations if needed. I don't need the level of control but it feels good to know that I have it if I need it.

Review from Matthew Turner

Matthew Turner

Review Origin Canada
5 stars review

Financially wise decision

Why install Chatwoot on a VPS that will cost me $20-25/month. Then worry about security, updates, backups, etc. Instead, I can activate it on DANIAN the same day for half the price. Plus, less headaches for maintenance. So happy I found out about this company!

Review from Andreas Schmitt

Andreas Schmitt

Review Origin Germany
5 stars review

Great service!

I configure custom Matomo tracking for my clients as a monthly service. Many of my clients were asking if they can selfhost Matomo instead of using the Cloud service. My employee recommended DANIAN. Basically, I can activate as much Matomo's in any of 20+ countries the same day. Additionally the affiliate revenue is a nice bonus.

Review from Jacob Turner

Jacob Turner

Review Origin United States
5 stars review

Smooth experience

I was listing through youtube tuts on how to install Bookstack myself. Then randomly, I found their video playlist exactly for Bookstack. Got curious and went at their website. For the same price (or higher) that I would pay for just the VPS, with them I'm getting a fully managed Bookstack. It's a no-brainer decision!

Review from Elias Gustafsson

Elias Gustafsson

Review Origin Sweden
5 stars review

I love opensource software

2 days ago I haven't heard about opensource softwares. Today I'm running 4 apps and enjoying the trip. Hey, 1 month later I may become a developer, haha just kidding. Love you guys!

Review from Mason Hayes

Mason Hayes

Review Origin United Kingdom
5 stars review

Its a great service

Its a great service. I love the performance of my Mastodon community. It keeps on growing and I don't have any worries about backups, secuirity, updates, etc.

Review from Vinícius Rocha

Vinícius Rocha

Review Origin Brazil
5 stars review

Very intuitive solution!

I wanted to use a nocode software for my company. A friend recommended them so I tried both Baserow and NocoDB, couple days later picked one.

Review from Haruto Hashimoto

Haruto Hashimoto

Review Origin Japan
5 stars review

Top-notch security

Honestly, I don't care much about performance and functionalities. I just wish to not bother with hacking attacks so that I can focus on my firm. I recommend them!

Review from Henrik Otto

Henrik Otto

Review Origin Germany
See more reviews

USE CASES

Three teams who run Qdrant on DANIAN

Vector search rarely arrives on its own. It turns up next to an application, an automation tool or a document pipeline, which is why the region choice and the payload design matter more than raw index speed.

12-PERSON SAAS TEAM

Adding an assistant without handing retrieval to a meter

The product ships an in-app assistant over customer documents. One collection, partitioned by an indexed tenant field rather than a collection per customer, with JWT tokens scoping each key. The instance runs in the same region as the application servers and is queried over gRPC from the backend.

LEGAL RESEARCH PRODUCT

Filtered retrieval across a document corpus that keeps growing

Filings and decisions are embedded once and queried with dense and sparse vectors together, filtered by jurisdiction and filing date through payload indexes. Scalar quantization keeps memory down without the recall loss binary would cost. Snapshots are pulled before every re-index.

3-PERSON AUTOMATION SHOP

A notes assistant wired together in an evening

The n8n instance and the Qdrant instance sit on the same account, both at €9. Documents arrive through a workflow, get embedded by an external model, and land in a single collection. Nobody on the team writes Rust, reads a config file, or holds a server.

COMPARISON

Four ways to run Qdrant

Vector databases are not priced per seat, so the honest comparison is not about headcount. It is about what the meter counts, what the hardware costs, and who gets paged when the optimiser stalls at three in the morning.

Path1 person5 people10 peopleWhat you also carry
Proprietary SaaS (Pinecone)$50/mo minimum$50/mo minimum$50/mo minimumThe bill tracks stored vectors and read units, not people. Storage, writes and reads are metered on top of the minimum, so a growing corpus raises the floor whether or not anyone new joins.
Self-host on a VPS~$40/mo~$40/mo~$40/moAn 8 GB box covers a million 768-dimension vectors. You own patching, snapshots, key management, firewalling and the upgrade path, plus a few hours a month that do not appear on the invoice.
Home server~$600 once + power~$600 once + power~$600 once + powerA four-bay box such as a Synology DS923+ holds the index in memory for development. Concurrent production query traffic, off-site backups and a residential uplink are where it stops being viable.
DANIAN managed Qdrant€9/mo€9/mo€9/moFlat per app, in the region you choose. Backups, certificates, monitoring, patching and support are inside the price. Resources scale on request and we do not change them without your say-so.

Query volume, not team size, is what moves a vector-database bill on the metered paths. That is the reason the flat line is worth something here: a retrieval layer that gets busier does not get more expensive.

BY INDUSTRY

Qdrant for specific industries

Retrieval quality is domain-shaped. A patent search fails differently from a product search, and the settings that fix one make the other worse, so these four sectors get different collection designs rather than the same default.

Client-file retention duties under professional-practice rules mean a matter's documents have to remain retrievable years after the work closes, and patent docketing carries its own record-keeping deadlines. That makes the corpus permanent and always growing.

We configure these instances with scalar rather than binary quantization, because the recall cost of binary shows up exactly where prior-art search cannot afford it, and we schedule a snapshot before every re-index so a bad embedding run is reversible. Payload indexes on jurisdiction and filing date turn the filter into part of the search rather than a pass afterwards.

The scale is real: one published patent-retrieval deployment covers hundreds of millions of documents from international patent offices and reports sub-second responses at a billion vectors.
Marketplace product-feed specifications dictate how catalogue attributes are structured and how often they are refreshed, so the vector layer has to accept a full re-upload on a schedule rather than a slow trickle of updates.

We run those refreshes with the indexing threshold dropped to zero and restore it once the load finishes, which is what keeps the optimiser from fighting a nightly bulk push. Payload indexes on category, price band and stock state are created before the first upload, not after, so the index is not rebuilt on every batch.

Semantic discovery and item-to-item recommendation run off the same collection. One published review-platform deployment holds 2.7 billion review vectors on this engine.
Support desks work to response-time commitments in their service agreements, and a retrieval assistant that answers slowly is worse than none, because the agent waits for it before answering the customer.

We issue a read-only API key for the retrieval path so the assistant can query but never write, and keep payload on disk so memory tracks the vector count rather than the size of the manuals. Filters on product line and release channel stop an answer from an old version reaching a current customer.

The gains are measurable. One published deployment scaled past five thousand connected sources, cut vector storage fourfold with scalar quantization, and moved queries from five to ten seconds down to under one.
Clinical record-retention rules and data-residency requirements decide where the index may physically sit, which for many providers means records must stay at rest inside one named country rather than wherever a vendor has capacity.

Region choice is set at creation on our side, so the instance runs in the jurisdiction the provider names and stays there. TLS and the API key are configured before first login, and the dashboard port is never open to the internet. Backups land in a second datacenter that you nominate.

Sizing is predictable: two million de-identified note embeddings at 768 dimensions need roughly 9 GB of memory, or about 2.3 GB with scalar quantization enabled.

FAQ

Frequently asked questions

Three groups of questions. Configuration and index behaviour first, then what a move from another vector service actually involves, then the platform questions about billing, backups and support that apply to every app we run.

01

Technical and configuration

No. A default build binds to every network interface with no API key configured, which is how instances end up publicly readable. We set the API key and the read-only key on first boot and keep the ports behind the firewall, so the instance is never reachable unauthenticated.
Qdrant sizes memory as vectors multiplied by dimensions multiplied by four bytes, then about 1.5 times that for index and metadata overhead. One million 768-dimension vectors lands near 4.6 GB. Tell us the vector count and dimension and we size the instance before you upload.
Yes. Scalar quantization cuts vector memory roughly fourfold and binary roughly thirty-twofold, with a recall cost that depends on your embedding model. We leave it off by default and turn it on with you after your own recall testing, per collection rather than globally.
Payload stays on disk by default and only the payload indexes stay resident, so memory tracks vector count rather than document size. Create the payload indexes you filter on before the bulk upload, otherwise the index build repeats after every batch.
Yes. It rides the same REST port as the API, which is exactly why that port is not open to the internet on our side. You reach the dashboard through your app URL with the key set, and we would not expose it directly on any host.
REST is easier to debug and is what the dashboard and most tutorials assume. gRPC is faster for batch upserts and heavy query volume. Both are enabled on separate ports and every official client library switches between them with one setting.
Yes. Dense vectors, sparse vectors and multivector late-interaction models run in the same collection and the same query, fused with reciprocal rank fusion or distribution-based score fusion. None of it is held back for the vendor's own hosted tier.
A single node handles most production workloads up to the memory ceiling of the plan. Clustering earns its complexity when you need a replication factor above one, which is what allows an upgrade to roll through without taking search offline.

02

Migration and onboarding

21 datacenter locations across six continents. You choose the region at provisioning. Application data sits in the region you choose; pick whichever is closest to your users or matches your data-residency preference.
Yes. Request a region migration from the dashboard and we run the move in the background. The system emails you when the migration completes; total transfer time depends on data volume but typical instances finish in a few hours. There is no extra charge for a region change.
Yes. Full data export is available at any time, in a portable format you can bring to any infrastructure.
The project publishes a migration tool that streams points from Pinecone, Weaviate, Chroma, Milvus and other Qdrant instances in resumable batches. You point it at the source and at your instance.
No, as long as the source vectors keep the same model and dimension count, because the engine stores whatever you send. Re-embedding is only needed when you are changing the embedding model, which is a good moment to move to a fresh collection name.
The instance is running before you log in. You get the URL, the API key and the dashboard. Most teams create their first collection and push a test batch within the hour, then wire the client library into the application.

03

Billing, support, and platform

€9 covers everything we do for that app: hardware in the region you choose, daily off-site backups with one-click restore, automatic security patches and version upgrades, 24/7 monitoring, SSL and firewall, and engineering support on email and the ticket section in the client portal. There are no setup fees or hidden line items. For more info see our Pricing page.
If you decide to continue, we charge €9/app/month from day 8. If you don't, the trial ends and you can export your data. No card is required for the trial, and we never auto-charge you without explicit consent.
No. The €9/month is flat regardless of how many users log into your app. Add 5 users or 50; the price doesn't change.
24/7 email and ticket support, both staffed by engineers who run the systems. We handle DNS configuration, SMTP setup, app integrations, performance tuning, troubleshooting, and migration help. Response time is typically under an hour. There is no tier system — every customer gets the same support.
Yes. Cancel from the dashboard. We don't charge a cancellation fee, we don't lock data, and we will export your data to you on request before deletion. data to you on request before deletion.
Every customer instance is backed up daily to a separate region from the primary. We test restores. You can request a restore at any backup point within the retention window — usually 7 days for daily backups.
Your application data sits in the region you choose at provisioning — 21 datacenter locations across six continents. Account-level data (billing, account email, support ticket history) is processed centrally. Application data region is picked by you, per app.
99.9% uptime SLA on every app, every tenant. Service credits are documented at danian.co/service-level-agreement. The status page is located at status.danian.co.
When your tenant approaches the resource ceiling — the base tier holds 1 vCPU/RAM, 30 GB storage — we notify you. Resource upgrades happen with your explicit consent; we will not upgrade your tenant or charge you without it.
We wait. We don't suspend the app or delete your data on the first failed charge. We email you, you fix the card on file, and we continue.
Invoices can be downloaded from the billing dashboard in PDF the day each charge succeeds. EU VAT is added where applicable and the VAT-reverse-charge regime applies for VAT-registered businesses with a valid number.
150+ open-source apps across automation, team chat, file sync, analytics, AI, password management, email marketing, dev tools, project management, smart home, CMS, and federated social. See the full catalog →
Yes. Every instance comes with a web-based terminal and a file manager in your DANIAN management dashboard. Useful for managing your data and customizations.
Resources scale with your usage. If your app needs more vCPU, RAM, or storage, we add it — and we ask first before any change to your plan. €9 is the floor; resource-heavy workloads may price higher, but you'll always know in advance.
Yes. We have both a Partner program and an Affiliate program available. Anybody can sign up.
No contract. No minimum commitment. Cancel anytime from the dashboard with one click. The 7-day free trial requires no credit card. After the trial converts to paid, you can still cancel at any month without notice or penalty.

DEPLOY IN YOUR REGION

21 datacenter locations on six continents

Pick the region closest to your users.

United States, Germany, Finland, Singapore, Australia, Brazil, Canada, Netherlands, UK, Spain, Italy, France, Sweden, Malaysia, India, Japan, Mexico, Poland, South Korea, Chile, South Africa and more coming soon

Global Reach Map
Browse all 150+ apps →

Try managed Qdrant for 7 days

No card. Cancel from the dashboard.
Start free trial