DotCheck

We train detection models, measure them on holdouts we publish, and ship the product that runs them.

This Hub org is the open side of that work: Apache-2.0 open weights (.npz heads on frozen upstream backbones), model cards, and the same gate numbers we use in production. Scores are estimates, not verdicts. The tables are what we measured. Use them as you will.

See what's real.

Hub id	Wire	Backbone / stack	Artifact
vermeer-image-v14_2	inhouse@14	Niepce + Janssen pair stack	siglip2_base_patch16_224_linear_head_v14_2.npz
valla-text-v14_3	inhouse-text@14 (+ lang heads)	Rask + Comenius + named tinies; Ma Jianzhong zh	text_stack_head_v14_3_en.npz + lang
muybridge-video-v4_4	inhouse-video@4	Plateau + Blau pair stack; stamp-cap	siglip2_base_patch16_224_linear_head_v4_4.npz
helmholtz-audio-v3_1	inhouse-audio@3	Helmholtz trained head	dasheng_base_linear_head_audio_h3e.npz

• Collection: DotCheck Engines
• About: DotCheck/dotcheck
• Inventory: REGISTRY.md in this about repo / hub/REGISTRY.md in source

Product

Everyday scoring is the DotCheck product, not a Hub demo:

• Hub storefront (static Space): https://huggingface.co/spaces/DotCheck/see-whats-real
• Site: https://dotcheck.ai
• Guest Check: https://dotcheck.ai/check
• Pro API: https://dotcheck.ai/api (Authorization: Bearer dc_…)
• Method and gates: https://dotcheck.ai/docs · https://dotcheck.ai/how

Open weights and cards live here. Clip assembly that joins video frames with audio (Covenant) stays in the product path.

Holdout claims (SSOT)

Measured holdout floats live in product Data.json / MODEL-CHOICE.md (image bal 0.9924, EN text bal 0.979, video stamp-cap bal 0.9859, audio bal 0.995). Do not cite stale bake-off YAML.