themortgagellm™
Natural-language access to U.S. mortgage origination and performance data.
Enter to run · Shift+Enter for new line
Explore additional capabilities
Explore lead generation angles
Check Fannie + Freddie eligibility
Ask in a language other than English
Ask about GSE guidelines
Conduct market research
Benchmark a lender
Score loans before funding
Test LLPA pricing
Track loan performance post-funding
Assess and action rep & warranty exposure
Search Chats
Recent Chats
Loan-level scoring
Score a loan against our calibrated risk + pricing models. Pick a model, describe the loan, and get a calibrated probability with plain-English interpretation and a recommendation. Learn more about our scoring models →
1. Pick a model
2. Describe the loan
Advanced · paste / edit raw JSON payload
Edits to the structured form above push into this textarea; edits here go straight to the API on submit (the structured form is not re-read after you edit JSON).
About these models · AUC, training cohorts, calibration notes
- Credit Approval Probability — probability the application receives a lender approval decision (action_taken IN (1,2)). HMDA-style input. For LO triage + lead prioritization. Baseline ~65% (train) / 62% (OOT). 2018-2023 train, 2024-2025 test: AUC 0.94, ECE 0.0003.
- Credit Denial Probability — probability an application is denied for credit (action_taken=3). HMDA-style input. For pre-credit-pull screening + fair-lending self-assessment + counter-offer routing. Baseline ~16% (train) / 19% (OOT). 2018-2023 train, 2024-2025 test: AUC 0.91, ECE 0.0004.
- Pull-through — probability an application closes as an originated loan (action_taken=1). HMDA-style input. For pipeline-hedge sizing, lead pricing, capacity planning, lender benchmarking. Baseline pull-through ~62%. 2018-2023 train, 2024-2025 test: AUC 0.92, ECE 0.0004.
- Appraisal Waiver Probability (PIW / Value Acceptance / ACE) — probability a conventional conforming loan is granted an appraisal waiver instead of a full appraisal. GSE-input schema. Pre-DU/LPA heuristic. 2018-2023 train (22.4M loans), 2024-2025 test (2.6M): AUC 0.85, ECE 0.0008.
- Higher-Priced loan (HPML) — probability the loan prices into a Reg Z Higher-Priced Mortgage Loan (first-lien rate_spread ≥ 1.5pp or sub-lien ≥ 3.5pp or HOEPA). HMDA-style input. 2018-2023 train, 2024-2025 test: AUC 0.87, ECE 0.0003.
- Fannie vs Freddie channel choice — S-learner over the repurchase + 12/24/36-mo EPD models; predicts which GSE produces lower expected loss for this loan, with full delinquency curves.
- Repurchase risk (v4) — probability of rep-and-warranty repurchase (zbc 06/96). For post-funding QC + R&W reserve setting. 2013-2023 train, 2024-2025 test: AUC 0.72 (in-cohort random-split: 0.81).
- EPD 12-mo (v2) — probability of 60+ DQ within first 12 months. For pre-funding pricing / LLPA tier. 2013-2023 train, 2024 test: AUC 0.83.
- EPD 24-mo — same target extended to 24 months. For mid-life risk pricing + reserve setting. 2013-2023 train, 2024 test: AUC 0.78.
- EPD 36-mo — same target extended to 36 months. Captures full-cycle DQ risk including the refi-window peak. 2013-2022 train, 2023 test: AUC 0.75.
- Prepayment 12 / 24 / 36-mo — probability of prepayment (zbc 01) within first 12 / 24 / 36 months. For MSR valuation, pipeline lock-desk risk, fast-pay vs slow-pay pool composition. 2013-2023 GSE cohorts.
- GNMA EPD — government-insured EPD model (FHA / VA / USDA-RD / PIH). Different feature schema (use
agency, credit_score, ltv, ...). 2018-2023 train, 2024 test: AUC 0.76.