For the lone comp chemist — no HPC engineer required

The full CADD stack, operated in your AWS.

Hit-finding, docking, co-folding, MD, and FEP — the open-source CADD stack (GROMACS, OpenFE, Boltz-2, OpenFold3) run as one managed Slurm cluster inside your own AWS account. One queue, the right GPU on spot, your data never leaving your VPC. Not a pip install — a cluster operated for you.

3 design-partner slots · BYOC plans from $99/mo · free 30-day pilot

Hit-finding → FEP · one queue GROMACS · OpenFE · Boltz-2 OpenFold3 · Uni-Dock · AMBER BYOC — your data stays in your VPC
Why FEP

Docking ranks compounds. FEP ranks them right.

Docking, co-folding, and MM-GBSA rank fast for triage — but still swap rank order within a scaffold, enough noise to advance the wrong compound and burn $400K–1.5M finding out. For the lead-opt call, FEP tracks experiment to 0.38 kcal/mol on the TYK2 benchmark. The hard part was never the algorithm — it was the cluster.

FEP · MUE 0.38 experimental ΔΔG (kcal/mol) predicted ΔΔG
FEP — tracks experiment docking · co-fold · MM-GBSA — fast, noisy
See it run

From sbatch to ranked hits — watch a full campaign.

A 1,000-compound hit-discovery funnel — structure → dock → co-fold → MD, end-to-end in your AWS. One command. 73 minutes. $1.16.

 ·  EGFR ATP site (7TU3), Enamine Diverse 1K — one sbatch, unedited wall-clock.

The open-source CADD stack your comp chem team already cites

· · · · · · · · · · · ·
Time to a working CADD cluster

Cluster live today. First campaign in 30 days.

AWS PCS, DIY 4 – 8 weeks
HPC engineer hire + ramp 12 – 24 weeks
Clusterra today
What you get

A campaign workspace, not just a cluster.

A managed Slurm cluster in your AWS, the full CADD suite operated end-to-end, and every campaign's ΔΔG history accreting as your program's computational book-of-record.

01 / Campaign workspace

Every run builds your program's provenance record.

Perturbation networks, ΔΔG history, MBAR bundles, and per-job cost stamps accumulate in your own EFS — a cost-stamped record that lives in your account, not ours, and compounds into a system no competitor can pull out of your VPC.

02 / Full CADD suite

Hit finding to FEP — one cluster, one queue.

Structure prediction (OpenFold3), docking (Uni-Dock, AutoDock-GPU, GNINA), affinity rescoring (Boltz-2), MD (GROMACS, AMBER), and RBFE campaigns (OpenFE, FEP multidir) — the same job queue, Karpenter picks the hardware. No partition routing in your scripts.

03 / Your AWS, your IP

Structures, force fields, and ΔΔG data never leave your VPC.

Pre-IND structure data is program IP. It stays in your IAM perimeter — not on our infra, not routed through our control plane. BYOC means you see every dollar on your AWS bill, and your Savings Plans apply.

Three ways in

Console, CLI, or REST — same cluster, your call.

Submit from the web console if you'd rather not touch Slurm. SSH in and sbatch if you live in the terminal. Wire the REST API into your orchestrator if you automate everything.

01 / Web console

Campaign dashboard — submit, monitor, review.

Browse the CADD template catalog, fill the campaign form, hit submit. Live logs, per-job cost stamp, run history, ΔΔG records. For scientists who'd rather not write sbatch.

Launch the console →

02 / CLI

sbatch fep-campaign --network edges.csv

Slurm-native. SSH in and submit the way you would on any HPC system — same sbatch, squeue, sacct, scancel. For comp chemists who already live in the terminal.

CLI quickstart →

03 / REST API

POST /v1/jobs/submit

Wire Clusterra into your in-house pipeline, Airflow DAG, or Slack bot. JSON in, JSON out. For platform teams that want Clusterra as the compute backend, not the front door.

API reference →

Pick your workload

CADD first. Cryo-EM STA second. More after that.

The same managed Slurm cluster runs all three verticals. Start with the pain that's blocking you now and expand as your program grows.

/cadd

Drug discovery — hit finding, lead-opt, FEP campaigns

For the lone comp chemist at a Series A/B small-molecule biotech. OpenFold3 + Uni-Dock + Boltz-2 + GROMACS hit funnel in 73 min at $1.16. RBFE campaigns on the same cluster — no partition routing, no HPC hire.

Scope a CADD pilot →

/cryo-et

Cryo-electron tomography — RELION-5 STA, AreTomo, IsoNet

For structural-biology biotechs running RELION-5 / AreTomo / Warp·M / IsoNet / MemBrain pipelines. Multi-TB tilt-series stay in your VPC. CryoCloud covers SPA — we cover STA. No academic-cluster waitlist.

Open cryo-ET page →

/your-workflow

Something else — genomics, proteomics, custom pipelines

If you're running nf-core/sarek, AlphaDIA cohorts, MAPPs, single-cell, or a custom Slurm pipeline — the platform substrate is the same. New workloads onboard in days. Send us what you're running.

Email us →

Campaign provenance

Every run ships with an auditable receipt.

Not a log file. A tarball any reviewer can rerun end-to-end — with ΔΔG history, MBAR bundles, and per-job cost stamps accreting in your own EFS.

tar -tf campaign_provenance.tar.gz
campaign_provenance.tar.gz
├── manifest.json          # run metadata + hashes
├── perturbation_network.json # Lomap edge list
├── ddg_estimates.csv      # per-edge ΔΔG + MBAR
├── slurm.sbatch           # exact submit script
├── container.sha256       # image digest
├── input/
│   └── ligands.sdf / protein.pdb
├── output/
│   └── ranked_hits.tsv / ddg.csv
├── cost.json              # per-job AWS cost stamp
└── provenance.log         # signed, append-only
manifest.json
{
  "run_id":        "clst-21047",
  "git_sha":       "a3f9e21",
  "workflow":      "fep-campaign@0.3.0",
  "container":     "sha256:4c8a…d9f2",
  "submitted_at":  "2026-06-01T09:14:22Z",
  "edges":         24,
  "wall_clock_s":  18420,
  "cost_usd":      11.40,
  "nodes":         "g5.12xlarge × 1",
  "protocol_hash": "b91c…d47e",
  "signed_by":     "clusterra-prod"
}

Finance can expense it. A new scientist joining next year can clusterra rerun a 2026 campaign and get the same ΔΔG estimates. The SAR history your program accumulates becomes the switching cost that makes the cluster worth keeping.

Benchmarks

Full-pipeline cost on the CADD workloads you actually run.

Spot pricing, checkpoint-restart on, us-east-1. Reproducible benchmarks on public datasets — fork the repo, verify the numbers.

Workload
Cost
Reference
Hit-discovery funnel — structure → screen → affinity → MD
OpenFold3 + Uni-Dock 1K-cpd + Boltz-2 top-50 rescore + GROMACS 1ns MD · L40s spot · proof-of-concept scale
$1.16 73 min, 1,000 compounds
EGFR ATP site (7TU3) · Enamine Diverse 1K · OpenFold3 TM 0.984 vs crystal. Proof-of-concept screen — not representative of production FEP campaign cost.
Boltz-2 affinity rescoring
TCR-pMHC 5-chain co-fold · A10G GPU spot
$0.13 per complex
DockQ 0.59 reference set · post-cutoff DockQ 0.62 vs published crystals
OpenFE RBFE campaign — TYK2, 9-edge perturbation network
OpenFE 1.11.1 · 11 λ-windows · 5 ns production · A10G spot (g5.xlarge) · single replicate
~$100 9-edge campaign, n=3  ·  ~$11 per edge
n=3 replicates (accuracy-grade) · based on verified n=1 run of $33 (job 2958, June 2026) · TYK2 canonical benchmark, MUE 0.38 kcal/mol vs experiment · 24-edge campaign n=3: ~$265
AlphaDIA proteomics cohort
timsTOF Ultra 2 · HeLa 200ng · Mann lab library
$0.05 per sample, 13 min wall-clock
8,591 proteins, 105,414 precursors · 50-sample cohort runs in parallel, not serial
Sarek + Parabricks 30× WGS
Parabricks GPU alignment + DeepVariant · g5 spot
$2.59 per sample, 2h 32m
GIAB NA12878 · SNP F1=0.9966 / INDEL F1=0.9934 against hap.py truth set
Verified run · job 2958 · June 2026 OpenFE RBFE · TYK2 · 9-edge campaign · A10G spot (g5.xlarge)
n=1 verified run · 18 edge-legs · A10G spot (g5.xlarge)$33
n=3 (accuracy-grade, 3 replicates)~$100
Per edge, n=3~$11
24-edge campaign, n=3~$265
9-edge campaign, n=3 (quoted)~$100

n=1 run verified June 2026, job 2958, TYK2. Spot evictions mid-run; checkpoint-resume recovered every leg with zero lost work. n=3 is required for per-ligand ΔG estimates and error bars — that’s the number we quote. Every dollar lands on your AWS bill, not ours.

See all 12 benchmarks →

Where we fit

Operated BYOC for CADD — the gap nobody else occupies.

Every named competitor has a hard constraint that excludes them from the lone CADD scientist at a Series A/B biotech. Here is the honest map.

Capability Clusterra Rowan Schrödinger BioTeam AWS PCS
Full CADD suite: hit-find → FEP — operated end-to-end hit-find + docking, no BYOC FEP FEP+ only with Maestro license consulting, not SaaS substrate — you build it
Open tools (GROMACS / OpenFE — no proprietary force fields) proprietary TMD engine OPLS3e, Maestro
BYOC — compute and data in customer’s own AWS VPC tier is bespoke enterprise virtual cluster, enterprise contract
Simple plan price (no per-seat, no compute markup) BYOC plans $99–$1,499/mo — you pay AWS for compute, no markup $200 solo / $50K org / $25 per-cpd ~$100K+ / year project billing per-hr mgmt fee
One queue — any job gets the right hardware, no partition routing 10-partition cap, no cross-family
Campaign system-of-record (ΔΔG + cost history in your EFS) results on Rowan infra Maestro project files

Rowan wins where the browser UX matters and budget is tight at the per-seat level — it’s a great product for early exploration. Schrödinger wins where the Maestro + OPLS3e stack is already licensed. BioTeam wins where you need custom consulting. AWS PCS is the substrate we’re built on. Clusterra wins when you want the open-source CADD suite operated in your own AWS account at a price a Series A biotech can actually afford — BYOC plans from $99/mo, with compute billed directly by AWS at no markup.

How it works

FEP campaign running in 30 days.

Three steps. All in your AWS account.

01 / Connect

Connect your AWS account — about 10 minutes

BYOC via a cross-account IAM role. Your VPC, your billing, your data. We deploy the Slurm + Karpenter control plane into your account — nothing routes through ours.

02 / Pick a campaign

Pick a CADD template or bring your workflow

Hit-discovery funnel, RBFE campaign, GROMACS FEP multidir, AMBER REMD, OpenFold3 batch — or submit your own Slurm job. The cluster knows what hardware each step needs; you never write a partition name.

03 / Submit

Submit — results + provenance back in your EFS

sbatch, REST, or the console. Karpenter provisions the right instance family on Spot. Checkpoint-restart on eviction. Live cost meter. Per-job cost stamp. Campaign provenance bundle on every run.

Beyond raw AWS

AWS sells you the parts. We hand you the working instrument.

Batch, PCS, and ParallelCluster are those parts pre-sorted but still un-assembled. We deliver the cluster already built, validated, and running — in your own account.

01 / Day-one cluster

Working on day one — not a kit to assemble.

Batch, PCS, and ParallelCluster are kits: someone still has to design the cluster, build the containers, wire up Slurm, and tune the GPUs before a single job runs correctly. That’s a dedicated headcount’s worth of HPC/DevOps work. We hand it to you already built and validated — a managed Slurm cluster running right the day we turn it on.

02 / Survivable spot

Spot GPUs at ~60% off — without the dead runs.

Spot GPU is ~60–70% cheaper than on-demand, but on raw AWS a long-running job dies the instant the instance is reclaimed — so most teams never use it. We catch the 2-minute reclaim warning, drain the job cleanly, and resume it. The cheap compute becomes usable for the long jobs that actually need it.

03 / Costed & reproducible

Every run is costed and reproducible — automatically.

On raw AWS you get one monthly bill and no record of how a result was produced. We stamp every job with its dollar cost as it runs — spend visible per experiment, per user, per project — and capture the exact pinned container digest and script behind each run, so any result can be reproduced or re-run months later without rebuilding the environment.

All of that — the day-one cluster, survivable spot, and costed, reproducible runs — for less than a tenth of the hire it would take to build it yourself:

Pricing

Pick your plan by how much you run.

Every plan is the full platform. Plans differ only by how much management credit is included.

Solo
$99/mo per cluster

For one scientist getting started. Run real campaigns on a proper GPU cluster without committing to anything bigger.

Included credits$99/mo≈ 100 GPU-hrs
Rate$0.40/GPU-hr + $0.04/vCPU-hrmetered from your credit
OverageSame ratepast your included credits

Get started →
Most popular
Team
$499/mo per cluster

For a team running every week. Pay $499, get $999 of management credit — about 2× the value.

Included credits$999/mo≈ 1,000 GPU-hrs
Rate$0.40/GPU-hr + $0.04/vCPU-hrmetered from your credit
OverageSame ratepast your included credits

Get started →
Org
$1,499/mo

For running hard, with zero math. One flat price covers everything you run — it never changes.

Included creditsUnlimitedone flat fee
RateNoneno metering, ever
Overagenothing to meter

Talk to us →

Against any of those, a $99–$1,499/mo management fee is a rounding error — and not a license, a CRO, or a headcount you'd otherwise carry.

Questions, answered

What exactly am I paying Clusterra for?

For running your cluster — the managed software, the validated scientific stack, spot survival, and provenance. That is the only thing you pay us. The compute itself (GPUs, CPUs, storage) is billed separately and directly by AWS in your own account, at AWS prices — we never touch or mark up that bill.

Which plan should I pick, and how does it bill?

One scientist, light or occasional runs → Solo ($99/mo, includes $99 of management credit). Running most weeks → Team, the best value ($499/mo, includes $999 of credit — ~2×). Running hard and want one predictable bill → Org ($1,499/mo flat, unlimited, no metering).

“Management credit” is prepaid Clusterra fee — not compute, and not your AWS credits. Solo and Team meter it at one uniform rate, $0.40/GPU-hr + $0.04/vCPU-hr; once the credit is used up you keep paying that same rate, nothing changes. Hours vary by instance; AWS bills the compute separately and directly, at AWS prices.

Can I use my AWS & NVIDIA credits?

Yes — this is a real advantage of BYOC. Because compute runs in your own account, your AWS Activate and NVIDIA Inception credits pay for it directly. A SaaS platform that bills you cannot apply those credits — they'd expire unused. Many teams run for months on credits alone; the management fee is the only thing you pay us.

What's the pilot?

A free 30-day pilot: we stand up your cluster in your AWS, wire your workflows, and get a real campaign running with you — no charge.

Can I cancel anytime?

Yes — no lock-in. Cancel anytime and changes take effect on your next cycle. Because everything runs in your own AWS account, your data and clusters stay with you.

Start free — no credit card →

Or book a free pilot — we get a real campaign running, no charge.

$ sbatch fep-campaign --network edges.csv --target your_protein.pdb # Your AWS. Your data. Your ΔΔG history. 30-day pilot.

Scope a pilot on your target class.

30 minutes to scope. Free 30-day pilot with your first RBFE campaign running in your AWS account — no charge. Then BYOC plans from $99/mo. Three design-partner slots — closes when filled.

Prefer email? hello@clusterra.cloud · or join the community Slack →