MAGE

5-30 minutesPaperCodeNeeds Token/License

De novo antibody generation from antigen sequence using language models

Readiness

Readiness: Needs Token/License
Smoke: not_tested
Artifacts: not_applicable
Production: gated_huggingface_mage_weights_pending
Storage: <5
Strategy: shared PyTorch SIF + tool ext3 overlay; avoid per-tool torch duplication
Next: Obtain/approve access to gated Hugging Face perrywasdin/MAGE_V1 weights, then build/reuse CUDA PyTorch 2.1/Transformers/ProGen2 SIF and run one-sequence Slurm generation smoke before runner wiring.

No obvious large model/database dependency identified; validate upstream docs before install. 2026-07-05 SOP batch recheck: backend inventory found no staged MAGE source/weights/SIF/artifacts/runner. Upstream MAGE requires CUDA PyTorch/Transformers/ProGen2 and the perrywasdin/MAGE_V1 Hugging Face model is gated, requiring accepted access conditions before smoke. Reclassified from Smoke Pending to Needs Token/License.