openfold3-nim

Files (12)

• BENCHMARK.md
• SKILL.md
• skillspector-baseline.yml · config/
• evals.json · evals/
• trigger_evals.json · evals/
• api.md · references/
• examples.md · references/
• parameters.md · references/
• science.md · references/
• validation.md · references/
• skill-card.md
• skill.oms.sig

SKILL.md (first lines)

# OpenFold3 NIM

Predict biomolecular structures with OpenFold3. It supports proteins, DNA, RNA,
small-molecule ligands, and multi-entity assemblies. Use this guide for
basic hosted and local NIM use; load supplemental files only when the task needs
deeper context:

- `references/api.md`: exact endpoints, schemas, Docker flags, response fields.
- `references/science.md`: purpose, strengths, limitations, and model handoffs.
- `references/parameters.md`: molecule fields, MSAs, templates, samples, tuning.
- `references/validation.md`: artifact checks and scientific sanity checks.
- `references/examples.md`: compact hosted and local request patterns.

## Choose Mode

Ask only when context is unclear:

> Hosted NVIDIA API or local Docker NIM?

- Hosted URL: `https://health.api.nvidia.com/v1/biology/openfold/openfold3/predict`
- Local URL: `http://localhost:8000/biology/openfold/openfold3/predict`
- Local readiness: `http://localhost:8000/v1/health/ready`

Mode difference: the local prediction path has no `/v1/` prefix. Hosted requests use `Authorization: Bearer $NGC_API_KEY`. Supported local Docker
startup uses `NGC_API_KEY` (or `NVIDIA_API_KEY` via the preflight) for
registry login, entitlement checks, and first-run model downloads; pass it
into the container with `-e NGC_API_KEY`. Local inference requests use no
auth header after readiness, so bind the host port to loopback with
`-p 127.0.0.1:8000:8000`. Warm-cache key-free startup varies by image version
and should not be assumed.

## Auth And Environment

Use credentials already supplied in the environment or injected by a secret
manager. Do not load credential files, print keys, or enable shell tracing.
Confirm keys exist with shell tests.

Hosted needs `NGC_API_KEY` in the request header. Local startup needs
`NGC_API_KEY`, or `NVIDIA_API_KEY` as a fallback, plus `LOCAL_NIM_CACHE`.

## Local Docker

Use the official OpenFold3 NIM image and mount `LOCAL_NIM_CACHE` at
`/opt/nim/.cache`. Before executing setup, explain that registry authentication
sends the key to the NVIDIA registry at https://nvcr.io and first startup
downloads about 10–15 GB of model weights into the cache. Run deployment only
when requested; for a setup guide, provide the commands without running them.

When writing local setup commands, copy the preflight below exactly. Do not
replace it with a simple `: "${NGC_API_KEY:?Set NGC_API_KEY}"` check, do not
drop `NVIDIA_API_KEY`, and do not invent a default `LOCAL_NIM_CACHE`; those
lines are the repo's local NIM env contract. The default single-GPU launch
should show the literal `--gpus "device=0"`; choose a different device only
when the user asks.

```bash
set +x

if [ -z "${NGC_API_KEY:-}" ] && [ -n "${NVIDIA_API_KEY:-}" ]; then
  NGC_API_KEY="$NVIDIA_API_KEY"
fi
: "${NGC_API_KEY:?Set NGC_API_KEY or NVIDIA_API_KEY}"
export NGC_API_KEY
: "${LOCAL_NIM_CACHE:?Set LOCAL_NIM_CACHE}"

mkdir -p "${LOCAL_NIM_CACHE}"
chmod 755 "${LOCAL_NIM_CACHE}"

printf '%s\n' "$NGC_API_KEY" | \
  docker login nvcr.io --username '$oauthtoken' --password-stdin && \
docker run --rm --name openfold3 \
  --runtime=nvidia \
  --gpus "device=0" \
  --shm-size=16g \
  -e NGC_API_KEY \
  -v "${LOCAL_NIM_CACHE}:/opt/nim/.cache" \
  -p 127.0.0.1:8000:8000 \
  nvcr.io/nim/openfold/openfold3:latest
```

Readiness check:

```bash
until curl -sf http://localhost:8000/v1/health/ready; do sleep 5; done
```

## Request Pattern

Use `requests.post(..., json=payload, timeout=300)`. For local Docker tasks,
set `hosted = False` after the readiness check passes.

```python
import os
import requests

hosted = True
url = (
    "https://health.api.nvidia.com/v1/biology/openfold/openfold3/predict"
    if hosted
    else "http://localhost:8000/biology/openfold/openfold3/predict"
)
headers = {"Content-Type": "application/json"}
if hosted:
    headers["Authorization"] = f"Bearer {os.getenv('NGC_API_KEY')}"

seq = "MKTVRQERLKSIVR"
payload = {
    "inputs": [{
        "input_id": "prediction_1",
        "output_format": "pdb",
        "molecules": [{
            "type": "protein",
            "id": "A",
            "sequence": seq,
            "diffusion_samples": 1,
            "msa": {
                "main": {
                    "a3m": {
                        "alignment": f">query\n{seq}",

From NVIDIA/skills, Apache-2.0 AND CC-BY-4.0.

More from NVIDIA/skills