Tools · Computational Biology

Boltz-2

Accurate in silico screening for early-stage drug discovery

On this page

Prices, workflows, and method papers Open in the app

Run it from the API

Submit with Submit a job and the job_type below. Price it first with Estimate job reservation cost: submitting reserves that amount from your wallet, and the charge settles at the actual runtime.

Boltz-2 `boltz2`

Job type: boltz2
Hardware: t4l4a10 (default)l40sa100-40gba100-80gbh100h200b200
Typical runtime: 30 min on Nvidia A10

Payload

Provide at least one of: boltz_yaml, sequences.

Payload fields
Field	Type	Description
`boltz_yaml`	`string`	Boltz input YAML document (validated by boltz2-yaml.schema.json).
`sequences[]`	`any[]`	Optional structured sequences used to generate boltz_yaml when omitted.
`constraints[]`	`any[]`
`templates[]`	`any[]`
`properties[]`	`any[]`
`boltz_args`	`object`	CLI-style arguments for job_type=boltz2 (payload.boltz_args).
boltz_args.`recycling_steps`	`integer`	Default: `3`Limits: `≥ 0`
boltz_args.`sampling_steps`	`integer`	Default: `200`Limits: `≥ 0, ≤ 500`
boltz_args.`diffusion_samples`	`integer`	Default: `1`Limits: `≥ 1, ≤ 100`
boltz_args.`step_scale`	`number`	Must be strictly greater than 1 (legacy API check). Default: `1.638`Limits: `> 1`
boltz_args.`output_format`	`string`	Default: `"mmcif"`One of: `"pdb", "mmcif"`
boltz_args.`method`	`string \| null`	One of: `null, "md", "x-ray diffraction", "electron microscopy", "solution nmr", "solid-state nmr", "neutron diffraction", "electron crystallography", "fiber diffraction", "powder diffraction", "infrared spectroscopy", "fluorescence transfer", "epr", "theoretical model", "solution scattering", "other", "afdb", "boltz-1", "future1", "future2", "future3", "future4", "future5"`
boltz_args.`affinity_mw_correction`	`boolean`	Default: `false`
boltz_args.`sampling_steps_affinity`	`integer`	Default: `200`Limits: `≥ 0`
boltz_args.`diffusion_samples_affinity`	`integer`	Default: `5`Limits: `≥ 1`
boltz_args.`max_msa_seqs`	`integer`	Default: `8192`Limits: `≥ 1`
boltz_args.`subsample_msa`	`boolean`	Default: `false`
boltz_args.`num_subsampled_msa`	`integer`	Default: `1024`Limits: `≥ 1`
boltz_args.`use_msa_server`	`boolean`	Default: `false`
boltz_args.`msa_pairing_strategy`	`string`	Default: `"greedy"`
boltz_args.`use_potentials`	`boolean`	Default: `false`
boltz_args.`write_full_pae`	`boolean`	Default: `false`
boltz_args.`write_full_pde`	`boolean`	Default: `false`
boltz_args.`mmseqs_threads`	`integer`	Worker-only MSA helper; not a Boltz CLI flag. Limits: `≥ 1`

Example

from cognichem_client import CogniChem

client = CogniChem.from_env()  # reads COGNICHEM_API_KEY
payload = {
    "boltz_yaml": "sequences:\n  - protein:\n      id: A\n      sequence: MKFL\n      msa: \">query\\nMKFL\\n\"\n",
    "boltz_args": {"diffusion_samples": 1},
}

estimate = client.jobs.estimate(job_type="boltz2", payload=payload, resource="a10")
print(f"Reserves ${estimate.cost:.2f}")

job = client.jobs.submit(
    job_name="my-boltz2-run",
    job_type="boltz2",
    payload=payload,
    resource="a10",
)
status = client.jobs.wait(job.process_id)
if status.status == "completed":
    client.jobs.result(job.process_id, save_path=".")

curl -X POST "https://api.cognichem.com/api/v1/jobs/submit" \
  -H "X-Api-Key: $COGNICHEM_API_KEY" \
  -H "Idempotency-Key: $(uuidgen)" \
  -H "Content-Type: application/json" \
  -d '{
  "job_name": "my-boltz2-run",
  "job_type": "boltz2",
  "payload": {
    "boltz_yaml": "sequences:\n  - protein:\n      id: A\n      sequence: MKFL\n      msa: \">query\\nMKFL\\n\"\n",
    "boltz_args": {
      "diffusion_samples": 1
    }
  },
  "resource": "a10"
}'

{
  "boltz_yaml": "sequences:\n  - protein:\n      id: A\n      sequence: MKFL\n      msa: \">query\\nMKFL\\n\"\n",
  "boltz_args": {
    "diffusion_samples": 1
  }
}

Sample data from the job catalog; long values are shortened here. Each job_name must be unique among your jobs.

Workflow inputs

Molecules (list)SMILES
Protein structurePDB, CIF

Workflow outputs

ArchiveZIP
StructuresCIF

Run it from the API

Boltz-2 boltz2

Boltz-2 `boltz2`