Vertical Foundation Models
One frozen LFM2.5-350M backbone per vertical. A small modality encoder + LoRA + task heads. Four industries, one architecture: payments, motor insurance, Medicare claims, and SOC threat intelligence.
The recipe Liquid already ships in LFM2.5-Audio and LFM2.5-VL, applied to enterprise-grade structured data. Each vertical defines a small per-modality encoder that turns its native data (64 transactions, 128 telematics events, 128 Medicare claims, 64 security telemetry events) into pseudo-tokens the frozen LFM2.5-350M backbone processes natively with attention. Per-task LoRA adapters and task heads share the same backbone, so a new business objective is an adapter — not a new model build, not a new vendor procurement. The same backbone also retains its text-generation capability through zero extra parameters, letting analysts and underwriters ask natural-language questions about the very entities being scored. Four industries, four lightweight encoders, one shared 350M base.
4 specialist models
How It Works
One frozen LFM2.5-350M backbone per vertical.
A small modality encoder + per-task LoRA + task heads.
Medicare Claims: Three Payer Signals from One Model
US health payers run separate ML systems for care management, utilization forecasting, and SIU triage — each with its own actuarial vendor and refresh cadence. A single foundation model processes the full 128-event Medicare claim history with attention, surfacing admission risk, next-event forecasting, and claim integrity in under 50ms total. Adapters (~134 MB each) carry the trainable surface; the 350M backbone never moves. The same backbone answers natural-language questions about the beneficiary through a weight-tied text head — zero extra parameters. Demonstrated on CMS DE-SynPUF (public, deidentified); production deployment trains on the payer's own claims, inside the payer's VPC.
Motor Insurance: Crash Risk + Fraud + Renewal in One Pass
Motor insurers run 5-15 separate models across pricing, claims, and retention — each blind to signals that span tasks (a driver's braking pattern predicts both crash likelihood and renewal sensitivity). One PRAGMA event encoder turns raw telematics streams into backbone-native inputs. Three lightweight adapters share the same frozen LFM2.5-350M backbone and produce crash risk, fraud probability, and renewal lapse in under 50ms. A new objective — UBI tier, subrogation potential, repair fraud — is a new adapter, not a new model build. Underwriter Q&A from the same backbone, no separate NLP system.
Payments: Four Signals, One Forward Pass
Card networks maintain separate models for fraud, next-merchant, amount, and MCC enrichment — each its own training pipeline and inference cost. One transaction encoder turns 64 transactions × 15 features into 960 pseudo-tokens at d=1024, the frozen LFM2.5-350M backbone processes them with attention LoRA, and four task heads predict in parallel: fraud (BCE), next-merchant (CE over 10,003), amount bucket (CE over 16), and MCC (CE over 103). Approximately 16M trainable parameters on top of a 350M frozen base. Adding chargeback likelihood or decline-reason is a head, not a model.
SOC Threat Intelligence: Ten Signals from One Telemetry Window
Security operations centers run separate detectors for verdict, risk, attacker stage, next-tactic forecast, and reviewer signals — each its own ML pipeline and refresh cadence. One cyber telemetry encoder turns a 64-event window × 15 normalized fields into 960 pseudo-tokens, the frozen LFM2.5-350M backbone processes them with attention LoRA, and ten task heads predict in parallel: attack presence, risk, current stage, next tactic, next event type, response actions, identity compromise, lateral movement, exfil likelihood, and a benign-admin confounder. Trained on 61K real/semi-real anchors from OTRF, Splunk Attack Data, and LogHub with synthetic expansion. Adding persistence likelihood or a custom analyst label is a head, not a model.
Try each model
All Demos
Claims Foundation Model for Medicare
One model that reads a Medicare beneficiary's claim history and returns three payer signals together: admission risk, next-event forecasting, and claim integrity. Under 50 milliseconds, inside the payer's VPC, on the payer's own data.
Three payer signals from one model in under 50 milliseconds. Adding new ones takes days, not quarters.
Mobility Insurance Co-Pilot
One model that reads raw telematics data and returns three underwriting signals together: crash risk, claims fraud, and renewal retention. Under 50 milliseconds, on your own infrastructure, no policyholder data leaving your perimeter.
Three underwriting signals from one model in under 50 milliseconds. Adding new ones takes days, not quarters.
Transaction Foundation Model
One model that reads a 64-transaction history and returns four signals together: fraud probability, next merchant, amount range, and MCC enrichment. Alternate tabs run dispute triage, collections, and fraud-pattern attribution on the same model.
One encoder + frozen LFM2.5-350M + LoRA + four task heads in a single ~22ms forward pass on H100. A new SSL pretraining stage catches ~5× more fraud on newly onboarded merchants with almost no label history — shown on synthetic data.
SOC Threat Intelligence Foundation Model
One model that reads a 64-event security telemetry window and returns ten SOC signals together: attack verdict, risk score, current attacker stage, likely next tactic, identity compromise, lateral movement, exfil likelihood, and analyst response actions — in a single forward pass on H100.
One encoder + frozen LFM2.5-350M + LoRA + ten SOC heads: verdict, risk, stage, forecast, and reviewer signals predicted in a single ~10ms forward pass on H100.
Ready to deploy in your environment?