Model Registry¶
OpenMed ships a manifest-backed registry (openmed.core.model_registry.OPENMED_MODELS) that annotates every official checkpoint with metadata such as category, specialization, recommended confidence, Hugging Face IDs, device fit, and benchmark summaries. Use it to pick the right model, surface dropdowns in UIs, or validate incoming requests.
Exploring the registry¶
from openmed.core.model_registry import (
get_all_models,
list_model_categories,
get_models_by_category,
get_model_info,
get_model_suggestions,
)
print(list_model_categories())
oncology_models = get_models_by_category("Oncology")
for info in oncology_models:
print(info.display_name, info.model_id, info.recommended_confidence)
info = get_model_info("disease_detection_superclinical")
print(info.description, info.entity_types)
suggestions = get_model_suggestions("Metastatic breast cancer on paclitaxel.")
for model_key, info, reason in suggestions:
print(model_key, info.display_name, reason)
ModelInfoobjects includedisplay_name,category,entity_types, size hints, benchmark data, optional latency/RAM maps, optional recommended tier, auditedscript_coverage, and a default confidence threshold.get_model_suggestionsleans on lightweight heuristics to recommend models based on text snippets or hints (disease, pharma, oncology, etc.).
Metadata for UIs & validation¶
ModelInfo.size_category,.size_mb,.latency_ms,.peak_ram_mb, and.recommended_tierhelp you decide whether a model can fit on CPU-only infrastructure or a target device tier.entity_typesfeed dropdowns or filter chips in your frontend.recommended_confidencecan drive slider defaults or guardrails on API calls (pass it toanalyze_text).get_pii_models_by_languageexcludes any model whose audited tokenizer is explicitlyunsupportedfor a script claimed by that language. The underlying UNK, byte-fallback, and tokens-per-grapheme measurements remain available onModelInfo.script_coveragefor diagnostics and UI warnings.
Plan downloads before using data¶
Use models size to inspect download, disk, and estimated peak RAM requirements from the committed manifest. The default path is offline-safe and does not contact Hugging Face:
OPENMED_OFFLINE=1 openmed models size disease_detection_tiny
openmed models size --budget-mb 100
openmed models size --budget-mb 100 --format json
With a budget, the command lists only models whose remaining download fits and recommends the smallest qualifying snapshot for each task. Models already in the local Hugging Face cache are marked cached — 0 MB to download and count as zero against the budget.
Use --remote only when you explicitly want to refresh an estimate from the current Hub file metadata. Supplying an alias keeps that opt-in lookup focused:
Sizes use decimal megabytes (1 MB = 1,000,000 bytes). Remote inspection requires the optional openmed[hf] dependencies; ordinary offline estimates do not.
Keeping the registry fresh¶
If you add a new Hugging Face checkpoint, refresh models.jsonl and, when measurements are available, enrich it with benchmark and device-fit results. See Model Manifest for the schema and merge command.
Manifest rows drive openmed/core/model_registry.py; avoid hand-editing the registry for new models. A new row should include:
- The full HF model id (
OpenMed/...) and core release metadata. - Representative canonical entity labels.
- Required 11-script tokenizer coverage for PII-family entries, plus optional benchmark, latency, RAM, and recommended-tier enrichment.
- Per-format download sizes and per-script recall/leakage-floor evidence when the release model card advertises those decision fields.
CI will enforce type safety through the unit tests, and the docs automatically pick up the new entry via the examples above.