japanfold
Command-line client for JapanFold, a hosted API for protein structure prediction, binder design and protein embeddings. No GPU and no model weights on your side; the models run on JapanFold's servers in Japan.
pip install japanfold
export JAPANFOLD_API_KEY=jf_live_... # from https://japanfold.aiand.com/account/token
japanfold fold --sequence MKTAYIAKQRQISFVKSHFSRQ --wait --out ./out
It has no dependencies beyond the Python standard library (3.9+).
Key and endpoint
Every call needs a key. Sign in at japanfold.aiand.com/account/token
to get one; a new account comes with $100 of credit. The CLI reads it from --api-key, then
$JAPANFOLD_API_KEY, then ~/.config/japanfold/config.json (written by japanfold auth login).
The API lives at https://api.japanfold.aiand.com. Set JAPANFOLD_BASE_URL or --base-url to
point somewhere else.
Commands
japanfold models # every model and design protocol
japanfold fold --sequence MKTAYIAK... --model boltz2 --wait --out ./out
japanfold design spec.yaml --model boltzgen --protocol nanobody-anything --wait --out ./out
japanfold design --model rfd3 --protocol rfd3-binder --structure target.pdb \
--contig 'A1-150,60-80' --wait --out ./out
japanfold design --model pxdesign --protocol pxdesign-binder --structure target.pdb \
--chains A --binder-length 80 --hotspots A64,A65,A66 --wait --out ./out
japanfold embed --sequence MKTAYIAK... --model esmc-600m --wait --out ./out
japanfold wait JOB_ID --out ./out # wait for a job and download it
japanfold credits # balance left on the key
japanfold jobs list
--wait submits, waits for the job and downloads the results in one foreground command.
Without it a command prints the job id and returns; japanfold wait JOB_ID picks it up later
and is safe to re-run. fold is the same command as predict, and wait the same as download.
For agents
--json on any command prints machine-readable JSON on stdout and keeps human messages on
stderr. japanfold schema prints the OpenAPI 3.1 contract. Exit codes say what went wrong
without parsing text:
| Code | Meaning |
|---|---|
| 0 | success |
| 1 | job failed or generic error |
| 2 | missing or invalid key |
| 3 | invalid input |
| 4 | not found |
| 5 | rate-limited or at capacity |
| 6 | server unreachable |
Agents that prefer plain HTTP can skip the CLI: the docs cover the REST API, and the agent skill lives at github.com/japanfold/japanfold.
Metadata
Release files for japanfold 1.0.1
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| japanfold-1.0.1.tar.gz | 14.1 kB | Without provenance | Details for japanfold-1.0.1.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| japanfold-1.0.1-py3-none-any.whl | Python 3 | none | any | Details for japanfold-1.0.1-py3-none-any.whl |
Total release size: 28.6 kB
Release files / japanfold-1.0.1.tar.gz
| Download URL | japanfold-1.0.1.tar.gz |
|---|---|
| Size | 14.1 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
3659bbd2df5ff0eabafa871e7f18e388c514aeed6fc9174800e936ffdcaddc94
|
|
BLAKE2b-256 checksum How to use checksums |
7ec16e0f333cb1ada40b9226b1277701604891c88ba599e994e1badc0f7f16a2
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.16
|
Release files / japanfold-1.0.1-py3-none-any.whl
| Download URL | japanfold-1.0.1-py3-none-any.whl |
|---|---|
| Size | 14.5 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
fbb56da2eab8750540a312c0aede5decda85ec14b5f95912de49148986715986
|
|
BLAKE2b-256 checksum How to use checksums |
388e2b58725d18f51d071058a4c47a78d21c5f0ac7b85ce9c5b9a756e040c5a3
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.16
|
Security for japanfold 1.0.1
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.