Protenix v2

Enhanced open-source biomolecular structure prediction with Protenix v2

Docs

Input

Add a molecule to begin

Choose a building block to assemble your structure.

75 credits

Output

Configure input settings on the left, then click "Submit job"orLoad an example (it's free)

Human ubiquitin (76 residues)

Protein-ligand complex

What is Protenix v2?

Protenix v2 is ByteDance Research's enhanced open-source biomolecular structure prediction model, released under the Apache 2.0 license as part of the Protenix project. It predicts 3D structures of complexes containing proteins, DNA, RNA, small molecule ligands, and metal ions from sequence or structural inputs, all in a single run.

AlphaFold 3 itself has restricted access and prohibits commercial use. Protenix follows the same general all-atom structure-prediction direction with open-source code and Apache-licensed model parameters, making it a practical option for open-ended research and applications where AlphaFold 3's terms would be limiting. ProteinIQ now runs Protenix v2 by default while keeping v1, mini, and tiny checkpoints available for comparison or faster exploratory runs.

How to use Protenix v2 online

ProteinIQ runs Protenix v2 on GPU infrastructure with no local installation. Add one or more molecules, optionally configure seeds and model, then submit. Results arrive as downloadable CIF files ranked by confidence score, viewable directly in the 3D structure viewer. Each ranked prediction also includes the Protenix summary-confidence JSON, and you can optionally export atom-level confidence JSON for deeper inspection. MSA search runs automatically for protein inputs and takes a few minutes; disabling it significantly reduces runtime at some cost to accuracy.

Inputs

Each molecule is added as a separate entity. Up to 10 entries of each type are supported.

Input	Accepted formats
Protein	FASTA sequence, PDB file, or 4-letter RCSB PDB ID
Ligand	SMILES string, CCD code (e.g. `ATP`), SDF/MOL file, or PubChem compound ID
DNA	FASTA sequence (`A`, `T`, `G`, `C`)
RNA	FASTA sequence (`A`, `U`, `G`, `C`)
Ion	CCD code (e.g. `ZN`, `MG`, `CA`, `FE`)
Protein paired MSA	Optional uploaded or pasted `.a3m` alignment attached to a protein chain label
Protein unpaired MSA	Optional uploaded or pasted `.a3m` alignment attached to a protein chain label
RNA MSA	Optional uploaded or pasted `.a3m` alignment attached to an RNA chain label

For ligand files, Protenix requires a 3D conformer. 2D SDF files (flat structure with no z-coordinates) will fail at featurization.

Precomputed alignments attach to the chain letters shown on the main protein or RNA molecule cards. If there is only one eligible protein or RNA chain, you can leave the MSA target label blank and ProteinIQ will attach it automatically. Protein paired MSA and protein unpaired MSA are separate native inputs; RNA only supports unpaired RNA-MSA in this pass. Template files are still out of scope here.

Settings

Setting	Description
`Model seeds`	Number of random seeds (1-5, default 1). Each seed independently samples the diffusion trajectory and produces a distinct set of structures.
`Model`	Model variant. Protenix v2 is the default. See model options below.
`Use MSA`	Whether to run multiple sequence alignment for protein chains (default on). Adds 2-5 minutes but improves accuracy, especially for proteins with many homologs.
`Samples per seed`	Structures generated per seed (1-5, default 5). Higher values increase output diversity.
`Precision`	`BF16` (default, faster) or `FP32` (full precision). Negligible accuracy difference on A10G hardware.
`Training-Free Guidance`	Applies physical constraints (steric, torsion, distance) during diffusion sampling to improve geometric plausibility. Off by default; adds compute overhead.
`Export atom confidence JSON`	Saves the `full_data` JSON for each prediction. Useful for debugging or downstream analysis, but larger than the default summary metrics.

Model options

Model	Parameters	Training cutoff	When to use
Protenix v2	464M	2021-09-30	Default. Highest-accuracy model for most targets, with stronger antibody-antigen and ligand plausibility performance reported by ByteDance.
Protenix v1 base	368M	2021-09-30	Baseline v1 model for comparison and lower compute cost.
Protenix v1 base (2025 data)	368M	2025-06-30	Targets with templates deposited after 2021.
Protenix mini	135M	2021-09-30	Rapid screening, large sequences, or quick estimates before a full run. Roughly 40% faster.

Protenix v1 vs Protenix v2

Protenix v2 is the default model on ProteinIQ. It is a 464M-parameter model released in April 2026, with the same 2021-09-30 training cutoff used for fair comparison to earlier Protenix models.

The Protenix documentation reports stronger antibody-antigen performance and improved ligand plausibility compared with Protenix v1, while retaining MSA, RNA MSA, and template support. It also works with the Training-Free Guidance features introduced in the Protenix 2.0.0 package.

Protenix v1 remains available as a lower-compute comparison point. Its public checkpoints include the base 368M models, the 2025 cutoff model, and smaller mini/tiny variants.

How does Protenix work?

Protenix implements the AlphaFold 3 architecture, which replaces the iterative refinement of AlphaFold 2 with a diffusion-based structure generation process. Prediction starts from randomly initialized atom coordinates and progressively denoises them into a coherent structure over multiple cycles.

Multiple Sequence Alignment

For protein inputs, Protenix queries external MSA databases to find evolutionarily related sequences. Conserved residues and co-evolving residue pairs reveal spatial constraints that guide the structure prediction. The MSA step connects to an external server and is the primary source of runtime variability, typically adding 2-5 minutes per prediction.

If you already have alignments, ProteinIQ can use precomputed .a3m files instead of forcing a new search. Protein chains can receive separate paired and unpaired MSAs, while RNA chains can receive an unpaired RNA-MSA. These uploads are written to temporary files and referenced through the pairedMsaPath and unpairedMsaPath JSON fields used by Protenix.

Confidence estimation

Protenix outputs per-atom and per-residue confidence estimates derived from the same network that produces the structure. These are not validated against experimental data for each prediction; they reflect the model's internal consistency. A high-confidence prediction that matches an incorrect prior (e.g. a homolog with a different binding mode) may score well while being physically wrong.

Understanding the results

Protenix outputs ranked CIF files together with the native summary-confidence JSON for each prediction.

Metric	Scope	Interpretation
pLDDT	Per-residue, 0-100	>90: high confidence. 70-90: backbone likely correct. 50-70: low confidence. <50: likely disordered or unreliable.
pTM	Whole structure, 0-1	>0.5 suggests the overall fold is correct.
ipTM	Inter-chain interfaces, 0-1	>0.7 indicates a reliable interface prediction. Particularly relevant for protein-protein and protein-ligand complexes.

ProteinIQ also preserves the additional native metrics and flags from summary_confidence, including ranking_score, gpde, per-chain confidence summaries, pairwise interface scores, clash flags, disorder estimates, and the recycle count used for the prediction.

When running multiple seeds, structural agreement across the top-ranked predictions is a better indicator of reliability than any single confidence score. Divergent predictions across seeds suggest the structure is genuinely uncertain.

Protenix vs. alternatives

Several AlphaFold 3-style tools are available for biomolecular complex prediction.

Tool	Strengths	Limitations
Protenix	Open license, strong benchmark performance, supports SDF file inputs	Slower than Boltz-2; no covalent docking
Boltz-2	Fast, supports affinity prediction	Fewer input formats
Chai-1	Strong protein-ligand interface predictions	Restricted to non-commercial use
OpenFold 3	Open-source AF3 alternative	Earlier implementation, fewer features

For protein-only structure prediction without any ligands or nucleic acids, ESMFold runs in seconds with no MSA step. For targeted docking when a binding site is known, AutoDock Vina or DiffDock may be more appropriate than full complex prediction.

Related tools

Chai-1

Chai-1 is a multi-modal foundation model for molecular structure prediction. Predicts 3D structures for proteins, ligands, DNA, RNA, and multi-component complexes with high accuracy.

IntelliFold 2

Controllable biomolecular structure prediction model for proteins, ligands, DNA, RNA, and multi-component complexes. IntelliFold 2 supports fast v2-Flash inference, optional MSA generation, and ranked confidence outputs.

OpenFold-3

OpenFold-3 is an open-source AI model for biomolecular structure prediction, aiming to reproduce AlphaFold3. Predicts 3D structures for proteins, RNA, DNA, and small molecule ligands with high accuracy.

RosettaFold3

Open-source structure prediction neural network for proteins, nucleic acids, and small molecules. State-of-the-art accuracy with multi-chain support.

Boltz-2

Boltz-2 is a biomolecular foundation model for structure and binding affinity prediction. Supports proteins, ligands, DNA, and RNA in multi-component complexes. Automatically scales GPU resources for large complexes. Predicts binding affinity with near-FEP accuracy at 1000x faster speed.

LMI4Boltz

LMI4Boltz is a low-memory fork of Boltz for biomolecular structure and binding affinity prediction. It preserves Boltz inference behavior while reducing VRAM use with in-place pair updates, CPU offload, reduced precision pair representation, and aggressive chunking.

ABodyBuilder3

ABodyBuilder3 predicts antibody variable-domain structures from paired heavy and light chain sequences. It returns a PDB structure and, for the pLDDT checkpoint, per-residue confidence values.

AlphaFlow

Generate protein conformational ensembles with ESMFlow, the single-sequence AlphaFlow model family. Produces multiple diverse structures showing protein flexibility and dynamics.

AlphaFold2

AlphaFold2 via ColabFold for high-accuracy protein structure prediction. Uses MMSeqs2 API for MSA generation with no local databases required. Supports monomer and multimer prediction.

ESMfold

ESMfold is a fast, single-sequence protein structure predictor from Meta AI. Predicts 3D protein structures directly from amino acid sequences without requiring multiple sequence alignments (MSA), making it significantly faster than AlphaFold while automatically scaling GPU resources for larger proteins.

What is Protenix v2?