01
drip · for researchers
the unified peptide pipeline. one api. one vendor.
discovery, prediction, structure, simulation, scoring, and calibration ship as one call. delivered via licensed Aurora technology, priced per pipeline run, available to academic labs, biotech teams, and regulated industry.
aurora-licensed technology · drip-operated pipeline · output-only api
02
capabilities
drip vs the field
drip delivers the unified peptide pipeline via licensed aurora technology.
| axis | drip | cradle | biolm | chroma | esm3 | rfdiffusion | tamarind | menten |
|---|---|---|---|---|---|---|---|---|
| vendors needed to replicate drip's pipeline | 1 | 4 | 5 | 4 | 5 | 6 | 3 | 4 |
| unified pipeline (discovery → predict → fold → sim → score → calibrate) | ✓ | · | · | · | · | · | · | · |
| on-chain attestable results (uma oracle) | ✓ | · | · | · | · | · | · | · |
| customer-isolated prediction (inputs not retained or shared) | ✓ | · | · | · | · | · | · | ✓ |
| public api with pipeline-runs pricing | ✓ | enterprise only | ✓ | · | limited | self-host | ✓ | · |
replicating drip via a competitor stack: roughly $8k to $15k a month, plus six weeks of integration engineering. drip delivers the same outcome through one api call.
03
example
python client
installable from pypi at alpha launch.
from drip_peptides import DripClient
client = DripClient(api_key="...")
result = client.generate(
profile={"targets": ["anticancer"], "length_range": [8, 14]}
)04
pricing
priced per pipeline run
| tier | who | price | includes | quota |
|---|---|---|---|---|
| open research | academic labs, .edu, peer-reviewed publishers | free | discovery + predict only | rate-limited, attribution required |
| lab | small bio labs, postdoc-budget research | $499 / mo | full pipeline (discovery → fold → simulate → score) | 1,000 pipeline runs / mo |
| biotech | biotech teams, scaled research workloads | $2,499 / mo | full pipeline + customer-isolated prediction enabled | 10,000 runs / mo, priority queue |
| enterprise | regulated industry, dedicated workloads | $25k / mo + $25k setup | unlimited runs, dedicated infra, customer-isolated prediction, SLA | unlimited |
| desci | $DRIP-denominated, deferred post-launch | post-launch | prediction-market integrations + on-chain attestation | tba |
a pipeline run is one full call covering discovery, prediction, structure, simulation, and scoring. customer-isolated prediction means your inputs are not retained or used for shared model updates; this is not probability calibration in the post-hoc ML sense.
05
data handling
what we log and what we don't
we log the minimum metadata required for billing and abuse prevention: request timestamps, pipeline-run counts, tier identity, and error codes. this is what the invoice and rate-limiter run on. nothing else is retained at the account level.
sequence content submitted to predict, generate, fold, simulate, and score endpoints is not retained beyond job completion plus a 24-hour debug window. after that window closes, the submitted sequence is not recoverable from drip infrastructure. it is not written to any persistent store, not indexed, and not accessible to drip staff outside the debug window.
submitted sequences are not used for corpus updates, model retraining, or any shared model updates. the underlying peptide stack is licensed from Aurora; drip operates the inference pipeline and delivers output only. no customer input feeds back into the model. this applies at all tiers including customer-isolated prediction.
06
alpha access
request access
researcher access opens open-research-tier first. submit below; we'll reach out when your slot is live.
07
legal
research-use-only
- research-use-only (RUO). API output is not for therapeutic, diagnostic, or human-use applications without independent validation.
- output-only delivery. customers consume sequences, predictions, structures, simulations, scores, and calibration behavior via the API. no model access, no model deployment, no white-label, no embed, no fine-tuning of the underlying peptide stack.
- no resale, redistribution, or proxy access. API output is for direct customer consumption only. reselling output to a third party requires a separate agreement.
- the underlying peptide stack is licensed by drip from Aurora. drip operates the pipeline and the public API; Aurora owns the model. citation path: the Aurora model card and Zenodo DOI on /methodology.
- OpenMTA-compatible. material transfer follows the OpenMTA framework when wet-lab handoff occurs.
- ORCID is optional but recommended. it accelerates open-research-tier verification.
- customer-isolated prediction engagements run under bespoke agreements. customer-isolated prediction means your inputs are not retained or used for shared model updates, not a model handle: contact us before signing.