
CardIndex
Collectr
CollX
Ripix.cards
Cardstream.ai is open-source software that identifies trading cards the moment they appear on a live stream: name, set, card number and confidence in about half a second. It runs on your CPU and calls the recognition API once per card, not per frame.

NovoPsych
Clinical assessment tools for psychologists, therapists and occupational therapists. Send measures by secure link and review scored results, severity bands and reports.

Which is more popular?
Website, pricing, platforms and company facts side by side.
|
|
|
|
|---|---|---|
| Website | cardstream.ai | clientforms.app |
| Pricing | ||
| Company | Startup from the Czech Republic · 2026 | — |
| Listed in |
In their own words, as submitted to SaaSHub.


Cardstream.ai is open-source software that identifies trading cards on live streams. Point a camera at a card and it returns the name, set, set code, card number, series and year in about half a second, with a confidence level and up to four alternative matches. It supports 25+ trading card...
Clinical assessment tools for psychologists, therapists and occupational therapists. Send measures such as the PHQ-9, GAD-7 and DASS-21 by secure link and get scored results, severity bands and reports ready to review. The catalogue also includes DSM-5 ADHD assessments and autism screening tools....
What each product offers, as listed by its team.


An editorial look at what each product does well and who it suits.


No analysis of CardStream AI yet.
Overall verdict
Why this product is good
Recommended for
Walkthroughs and reviews on video.
Cardstream Identifying a Card in Real Time
No ClientForms.app videos yet. You could help us improve this page by suggesting one.
How often each product is chosen within a category, 0–100% relative to the other.


As answered by people managing CardStream AI and ClientForms.app.
CardStream AI's answer
Python, ONNX Runtime, an RF-DETR segmentation model trained by Ximilar for finding the card, and an embedding model for the same-card check (Torch, ONNX or LiteRT backends). It also uses ffmpeg for RTSP, RTMP and SRT streams, a local browser interface, Docker, and Ximilar’s Collectibles Recognition API for identification.
ClientForms.app's answer:
Frontend: Next.js 15 (React 19), Tailwind CSS Backend: Supabase (PostgreSQL, Row-Level Security) Authentication: Clerk Payments: Stripe Hosting: Vercel (Sydney region) Database hosting: Supabase (Australian data residency)
CardStream AI's answer
Card breakers and live sellers on Whatnot, Fanatics Live, eBay Live and TikTok Shop, and card shops streaming on Twitch or YouTube. It’s also for developers who want to add card recognition to their own streaming or scanning setup.
ClientForms.app's answer:
Psychologists who assess ADHD and autism. Typically private practice, solo or small group (1-5 clinicians), currently using paper forms and spending hours on admin per assessment. They want something that works and is secure.
CardStream AI's answer
Cardstream.ai is open source and runs on your own computer. It watches the video you already stream and identifies each trading card in about half a second, with no button to press. A local engine checks whether the card has stopped moving and whether it’s a new card. Only a new card triggers a paid identification call: about 100 calls an hour instead of 54,000.
ClientForms.app's answer:
Neurodivergent-first design. Most digital forms just replicate paper online. ClientForms is built specifically for ADHD and autism patients - progress bars with time remaining (helps with time blindness), softer colours instead of harsh white, sticky headers so you always know where you are. The forms are designed to reduce anxiety and improve completion rates.
CardStream AI's answer
Most card recognition tools for live streams are either closed paid services or browser extensions that identify one card per click. Cardstream.ai is free and open source (Apache-2.0), works automatically on the live video, and needs no GPU. It never logs in to your marketplace accounts. Only one image crop per new card leaves your machine, and you can connect your own recogniser instead of Ximilar’s API.
ClientForms.app's answer:
Most clinical software treats patients as an afterthought. ClientForms is designed specifically for neurodivergent users who are filling out the forms, progress indicators, time estimates, calmer colours, predictable layouts. One link, no signup required, works on any device. Patients just open and complete.
CardStream AI's answer
Cardstream.ai is built by Ximilar, a computer vision company in Brno, Czech Republic, and the maker of the Collectibles Recognition API. In live card breaks the seller opens packs on camera and names every card by hand. Cardstream.ai identifies those cards from the stream automatically. Most of the engineering went into keeping it cheap to run, so that the paid API is called only when a new card appears. It was first released as open source in September 2026.
CardStream AI's answer
Leave this empty. There are no named public users yet, and a made-up list would be easy to disprove.
Share your experience with using CardStream AI and ClientForms.app. For example, how are they different and which one is better?
When comparing CardStream AI and ClientForms.app, you can also consider the following products.

Real-time Pokémon card prices and sales data across major marketplaces
Compare CardIndex to CardStream AI or ClientForms.app:

NovoPsych is a tool that is used for outcome monitoring via conducting psychological tests online.
Compare NovoPsych to CardStream AI or ClientForms.app:

Collectr is the world’s fastest growing collectibles portfolio manager.
Compare Collectr to CardStream AI or ClientForms.app:

CollX (“collects”) answers the age-old question: “What’s it worth?” The app can scan any baseball, football, or basketball card and instantly ID it and get the avg market value.
Compare CollX to CardStream AI or ClientForms.app:

AI-powered sports card scanner that identifies cards and prices them from real eBay sold data.
Compare Ripix.cards to CardStream AI or ClientForms.app: