Personal project

A four-model AI pipeline for trading card collection.

Every card passes through four AI models before it gets a valuation — three running fully local on a GB10 AI PC, and a fourth handled by Cade, my card specialist agent on a completely separate machine using ChatGPT 5.5. The identification stack alone totals nearly 100GB of models. This is what happens when you stop using AI as a tool and start designing it as infrastructure.

View card inventory ↗
2023 Topps Finest Mike Trout Oil Spill Refractor PSA 10 — front
2023 Topps Finest Mike Trout Oil Spill Refractor PSA 10 — back

Mike Trout

2023 Topps Finest · #101

Auto Oil Spill Refractor PSA 10 Auto-flagged ⚑
Low $11.29
Median $74.99
High $149.99
Cade $150.00

"Pricing at the high end of recent eBay sales due to the scarcity and desirability of Mike Trout's Oil Spill Refractor in Topps Finest, graded PSA 10."

How it works

Four models. Multiple cards at once.

Most collectors use one app that queries a single price database. This pipeline runs four specialized AI models across two machines — up to 4 concurrent workers on the GB10 for identification, 8 parallel workers on the Lenovo for comps — processing multiple cards simultaneously at every stage.

Model Size Role Runs on Concurrency
llama3.2-vision:11b ~12 GB Phase 1 & 2 visual scan — front and back GB10 (local) Up to 4 concurrent
qwen2.5:32b ~20 GB Text reader — serial numbers, fine print, set IDs GB10 (local) Concurrent with 11b
llama3.2-vision:90b ~62 GB Deep ID — flagged cards only, full checklist prompt GB10 (local) One at a time, exclusive
ChatGPT 5.5 Cade — comparable sales, market movement, valuation Lenovo (separate system) 8 parallel workers
01
iPhone App

Scan from anywhere.

A custom iPhone app I built lets me scan cards from anywhere — at a show, at home, in a box of cards I haven't touched in years. The image hits my GB10 over the local network and lands in an intake folder. The pipeline wakes up automatically.

Custom iOS app

Network transfer → GB10
02
llama3.2-vision:11b — Phase 1 & 2

Fast visual scan — front and back.

llama3.2-vision:11b (~12 GB) runs on every card, up to 4 at once. It reads the front for slab detection, serial numbers, parallel finish, autograph presence, and patch — then flips to the back for player name, year, brand, set, and card number. Fast and parallel by design.

Parallel, patch, signature, and numbered cards are automatically flagged here for the 90b deep pass.

GB10 · 11b · ×4

02b
qwen2.5:32b — concurrent

Text reader runs at the same time.

While the vision model scans the image, qwen2.5:32b (~20 GB) runs concurrently to extract fine-print text — serial numbers, edition lines, copyright strings, and set identifiers that a vision model reads but may not interpret precisely. Two models, one card, simultaneous.

GB10 · 32b

Flagged cards only — exclusive, one at a time
03
llama3.2-vision:90b — flagged cards only

Deep identification. One card at a time.

llama3.2-vision:90b (~62 GB) runs exclusively on flagged cards — parallels, graded slabs, numbered prints, signed cards, patch relics. It uses a full checklist-constrained prompt loaded with known parallel names for the specific set, so it doesn't guess — it matches. This model runs one card at a time with no concurrency. It earns that exclusivity.

GB10 · 90b · ×1

Identified cards → watched folder → Lenovo picks up
04
Cade — ChatGPT 5.5 on Lenovo

Comps on a completely separate machine.

Once all local models finish, the card data lands in a folder the Lenovo watches. When new entries appear, Cade triggers — a purpose-built card specialist agent running on ChatGPT 5.5. Cade pulls comparable sales, reads recent market movement, and returns a structured valuation with confidence context. Completely separate system, completely separate model. The GB10 and the Lenovo share nothing except a watched folder.

Lenovo · Cade · GPT-5.5

4 AI models per card
~94 GB local models on GB10
2 separate machines
3 local
+ Cade
identification then comps

Why run it locally

Speed, cost, and control.

Both systems run multiple workers simultaneously. On the GB10, llama3.2-vision:11b handles up to 4 cards concurrently alongside qwen2.5:32b for text — the 90b model gets exclusive single-card access when flagged cards need it. On the Lenovo, Cade runs 8 parallel workers, pulling comps on multiple cards at the same time.

Cade uses ChatGPT 5.5 for comps — market intelligence that requires current eBay sold data, live price movement, and condition-adjusted comparables. Identification stays local. Market knowledge goes to the model built for it.

Automatic flagging

Special cards get special handling.

The light visual model is trained to flag four card types the moment they're detected. Flagged cards are queued for the large image model immediately — they don't wait for the standard batch.

Parallel

Color variants and refractors — visually similar to base cards but with significant value differences. Early detection matters.

Patch

Relic cards with embedded jersey, patch, or equipment pieces. Patch quality directly affects comp value.

Signature

Autographs — whether on-card or stickered — require authentication context before any valuation is attempted.

Numbered

Print-run limited cards where the serial number directly determines scarcity tier and comparable pool.

Relic

Embedded memorabilia cards — jersey swatches, bat chips, equipment pieces — where relic type and size affect grade and value.

More from the collection

2025 Topps Signature Class Jaxson Dart Silver Auto /275 — front
Front
2025 Topps Signature Class Jaxson Dart Silver Auto /275 — back
Back
Auto Silver /275 Auto-flagged ⚑

2025 Topps Signature Class — Jaxson Dart #106. Flagged for autograph + numbered /275. Giants QB with Year 1 upside.

Low $400.00
Median $500.00
High $600.00
Cade recommends $425.00

"Raw/no slab seen. Exact Silver /275 Dart autos are thin but this is the Giants QB chase from a premium auto product; hold unless someone pays $550+."

Classic Baseball Bo Jackson green border — front
Front
Classic Baseball Bo Jackson green border — back
Back
Baseball

Classic Baseball — Bo Jackson. Auburn jersey, green border. Two-sport legend card from the Classic brand.

Low $10.00
Median $11.40
High $59.95
Cade recommends $60.00

"Pricing at the high end of recent comps to maximize potential profit given Bo Jackson's collectibility."

2023 Panini Impeccable Kevin Na Silver /30 — front
Front
2023 Panini Impeccable Kevin Na Silver /30 — back
Back
Silver Embedded /30 Auto-flagged ⚑

2023 Panini Impeccable — Kevin Na #26. Actual silver embedded in the card stock, numbered /30. Golf. Flagged for numbered + rare parallel.

Low $9.99
Median $27.50
High $244.68
Cade recommends $280.00

"Pricing at the high end due to its numbered /30 scarcity and the strong demand for parallels in golf sets."

2024 Panini Immaculate Collection LeBron James Gold /99 — front
Front
2024 Panini Immaculate Collection LeBron James Gold /99 — back
Back
Gold Parallel /99 Auto-flagged ⚑

2024 Panini Immaculate Collection — LeBron James SM-LBJ. Gold parallel, numbered /99. Flagged for parallel + numbered.

Low $118.99
Median $239.99
High $425.00
Cade recommends $450.00

"Pricing at the high end of recent comps due to the scarcity and desirability of the Gold parallel, numbered /99, for a hot player like LeBron James."

Current inventory

Every card that's been through the pipeline.

The live inventory tracks each card, its AI-generated identification data, Cade's comp, and current market range. It reads straight from the GB10 card database and updates as scans complete.

Open card inventory ↗

The iPhone app

Built for scanning anywhere, not just at a desk.

The iOS app was built from scratch to solve one specific problem: I needed to be able to scan a card at a show, in a trade situation, or out of a box in the garage and have the pipeline start running before I get home. The app sends the image directly to the GB10 over the local network and shows a status update when each model finishes.

There's no subscription. No card database API. Just a camera, a local network, and four models that know what they're looking at.

Card Vault scanner app running on iPhone — camera viewport, hint buttons, processing queue

Card Vault — PWA added to iPhone home screen

Scan any card from anywhere — the pipeline starts the moment the image lands on the GB10.

Contact handoff

For speaking, media, or collaboration inquiries, get in touch.

Reach out directly by email or phone. For speaking and media requests, include the event, audience, format, and date so the inquiry can be routed quickly.

A short note about topic, format, timing, and audience is usually enough.