What Tablif read on VerQen's site
VerQen describes itself as: “zorqelis-ai/VerQen · Hugging Face”
- Price: $1.7 to $4.0 per million · $1.81 per million
- Stage: Live, no waitlist
- How it differs: “It never writes free text.”
What it is
Calls itself “foundation decision model”.
- Calls itself: foundation decision model
“…zorqelis-ai/VerQen · Hugging Face”
“VerQen is a foundation decision model from ZorQelis AI”
Where it puts itself
Claims “best calibrated”.
- Superlative claim: best calibrated
“It never writes free text. It is a 232M-parameter encoder that ZorQelis AI trained from scratch”
“…best on banking77 (85.5%) and emotion (91.5%), best calibrated (ECE 0.039), 2 points behind Jev overall.”
What it costs
Prices quoted: $1.7 to $4.0 per million, $1.81 per million.
- Price: $1.7 to $4.0 per million · $1.81 per million
“$1.7 to $4.0 per million decisions on a cloud L4-class GPU”
“$1.81 per million”
Who it sells to
Sells to developers. In English.
- Sells to: developers
- Languages: English
“Guardrails for agents: which tool to call, whether an action is inside the allowed scope, whether a message is a prompt injection, when to stop and ask.”
“Languages other than English.”
How it runs
A API. Runs locally and works offline. No API. Open source. AI is the core of the product.
- Kind: api
- Runs where: on the device, offline
- API: no
- Open source: yes
- AI role: core to the product
“…pip install git+https://github.com/ZorQelis-AI/verqen”
“# downloads once, then runs locally”
What it works with
Names Jev, Laya and 3 more on its pages.
- Names: Jev · Laya · ZorQelis AI · +2
“Jev 1.13 (TypeSafe, API)”
“Laya (Convai)”
What it offers as proof
Live, no waitlist. Claims “22,864 cases”. Publishes docs.
- Number claimed: 22,864 cases
“Sign Up”
“22,864 cases in three tracks”
Sources
Tablif read these pages, last on 2 Oct 2026. Everything above is what the product says about itself; Tablif does not verify every claim.
Similar products in Tablif Town
- OpenDecider: Open decision models that route AI agents without LLM calls
- Strands Decider 2B: A small open source decision model for agents
- JevModel: Learn Jev. Run Choice, Score & Noul in your browser.
- Microsoft-Decision-1: Microsoft’s decision model for agents and workflows
- SelfJev: Jev-compatible self-hosted decisions mode
- DecisionsApi: Text in. Decisions out.
- Citra Decision System: Decision apps that learn your experts' judgement
- Verdant: Independent grow intelligence. Your data stays yours.