Hugging Face scored 5.6/10.

Hugging Face’s strongest dimension is Post-click.

Post-click scores 7.0 / 10. The dim covers 4 signals in the rubric; no findings landed against it on this page — clean execution.

Hugging Face vs the rest of ml hub + inference.

1. 01Replicate6.0/10
2. 02Hugging FaceYou are here5.6/10
3. 03Anthropic4.2/10

What’s costing Hugging Face, quoted from the page.

1. 01
   The hero button sends ready-to-sign-up visitors on a field trip instead.

   “Explore AI Apps”

   Your hero button is a browsing action — go look at things — but the people landing here already know the platform exists. The 'Sign Up' button doesn't appear until mid-page, after a full trending-models scroll. For visitors ready to commit, the page makes them wander before it offers a door.

2. 02
   The page has no signup prompt until mid-scroll, and when it finally appears, it's buried inside a feature list — not treated as a conversion moment.

   “The hero section has two CTAs: 'Explore AI Apps' (pointing at Spaces) and 'Browse 2M+ models' — both browsing actions. The 'Sign Up' button appears only in the 'Home of Machine Learning' section, well below the fold, styled as a secondary element inside a feature description bloc…”

   The page has no signup prompt until mid-scroll, and when it finally appears, it's buried inside a feature list — not treated as a conversion moment. A product-aware ML engineer who arrives ready to create an account has to hunt for the action.

3. 03
   The page gives enterprise buyers a feature list but no path to evaluate whether Hugging Face fits their procurement process.

   “The Team & Enterprise section shows 'Starting at $20/user/month' and lists SSO, Audit Logs, Resource Groups, Private Datasets — then stops. No CTA exists for enterprise evaluation: no demo request, no sales contact, no 'See Enterprise Plans' link. The only action available is the…”

   The page gives enterprise buyers a feature list but no path to evaluate whether Hugging Face fits their procurement process. There's no 'Talk to Sales' or 'Request a Demo' entry point — just a starting price and a feature checklist.

4. 04
   Pricing clarity gap: buyers want 'clear pricing,' page shows opaque GPU tiers

   “Starting at $0.60/hour for GPU”

   38% of buyers cite 'clear pricing' as a key requirement, yet the page surfaces only a starting floor ('Starting at $0.60/hour for GPU') with no ceiling, no usage examples, and no cost calculator — leaving the actual spend for production workloads unquantified.

5. 05
   Team and Enterprise pricing collapse into one unanchored block.

   “Team & Enterprise Give your team the most advanced platform to build AI with enterprise-grade security, access controls and dedicated support. Getting started Starting at $20/user/month”

   The page bundles Team and Enterprise into a single pricing block with one starting price, so a buyer evaluating an enterprise rollout can't tell what they'd actually pay or what they'd get beyond the Team tier. Without a visible split, the anchoring mechanism breaks — there's no contrast between a $20/seat team plan and whatever Enterprise costs, which is exactly the comparison a procurement team or engineering lead needs to bring a budget request internally.