Projects
View on GitHub
amorehead/awesome-generative-flows
amorehead/awesome-generative-flows has 51 GitHub stars and 6 forks. GitHot measured 0 stars in 24 hours, 0 in 7 days, and 0 in 15 days. Last checked 2026-11-14 23:35 UTC. Last push 2026-10-20.
Measured star growth
- in 24h
- 0
- in 7 days
- 0
- in 15 days
- 0
- in 1 month
- Collecting history
Net star changes over overlapping periods, not daily totals. Missing measurements are left blank. Stars measure attention, not software quality.
About
Awesome papers related to generative flow matching and its applications in bioinformatics. (Nature Machine Intelligence)
AIREADME.md
Original:robot: awesome-generative-flows
Awesome papers related to generative flow matching and its applications in bioinformatics.
Please feel free to create a pull request if you would like to add other awesome papers.
Contributors
Open-Source Methods for Flow Matching Models
| Method Name | Year | Brief Description | Source Code |
|---|---|---|---|
| Neural Autoregressive Flows (Huang et al., 2018) | 2018 | Foundational model | GitHub |
| Neural (latent) ODE (Chen et al., 2018) | 2018 | Foundational model | GitHub |
| GLOW (Kingma et al., 2018) | 2018 | Foundational generative flows | GitHub |
| FFJORD (Grathwohl et al., 2019) | 2019 | Improves computational efficiency of Neural ODE | GitHub |
| Augmented Neural ODEs (Dupont et al., 2019) | 2019 | Foundational neural ODE | GitHub |
| Residual Flows (Chen et al., 2019) | 2019 | Invertible generative flows | GitHub |
| Graph Normalizing Flows (Liu et al., 2019) | 2019 | Generative geometric flows | GitHub |
| Stochastic Normalizing Flows (Wu et al., 2020) | 2020 | Generative stochastic flows | GitHub |
| Equivariant Normalizing Flows (Garcia Satorras et al., 2021) | 2021 | Generative equivariant flows | GitHub |
| Diffusion Normalizing Flow (Zhang et al., 2021) | 2021 | Generative stochastic flows | GitHub |
| Monte Carlo Flows (Gabrie et al., 2022) | 2022 | Flows for MCMC | GitHub |
| Divergence-free Neural Conservation Laws (Richter et al., 2022) | 2022 | Flows with divergence-free neural networks | GitHub |
| Rectified Flows (Liu et al., 2022) | 2022 | Straight and fast flow trajectories | GitHub |
| Flow Matching for Generative Modeling (Lipman et al., 2022) | 2022 | Generative flows with Gaussian probability paths | GitHub |
| Stochastic Interpolants (InterFlow) (Albergo et al., 2022) | 2022 | Building flows with stochastic interpolants & diffusion | GitHub |
| RIVER (Davtyan et al., 2022) | 2022 | Flow matching for video prediction | GitHub |
| Point Straight Flow (Wu et al., 2022) | 2022 | Generating point clouds with straight and fast flows | GitHub |
| Action Matching (Neklyudov et al., 2023) | 2023 | Generative optimal transport with arbitrary paths | GitHub |
| OT Conditional Flow Matching (Tong et al., 2023) | 2023 | Generative (conditional) continuous flows | GitHub |
| Fast-ODE (Lee et al., 2023) | 2023 | Minimizing trajectory curvature of ODE-based generative models | GitHub |
| Latent Flow Matching (Dao et al., 2023) | 2023 | Flow matching in the latent space | GitHub |
| [SF]²M (Tong et al., 2023) | 2023 | Simulation‑free stochastic dynamics | GitHub |
| Riemannian Flow Matching (Chen et al., 2023) | 2023 | Generative geometric flows | GitHub |
| Equivariant Flow Matching (Klein et al., 2023) | 2023 | Generative equivariant flows | GitHub |
| Wasserstein Lagrangian Flows (Neklyudov et al., 2023) | 2023 | Unified flows and optimal transport | GitHub |
| FM Boosting (Fischer et al., 2023) | 2023 | Boosting latent diffusion models with flow matching | GitHub |
| Motion Flow Matching (Hu et al., 2023) | 2023 | Flow matching for human motion synthesis and editing | GitHub |
| InstaFlow (Liu et al., 2023) | 2023 | Flows for accelerating text-to-image generation | GitHub |
| Multimodal Flow Matching (Campbell et al., 2024) | 2024 | Continuous‑discrete data generation | GitHub |
| Functional Flow Matching (Kerrigan et al., 2024) | 2024 | Flow matching for infinite dimensional spaces | GitHub |
| Dirichlet Flow Matching (Stark et al., 2024) | 2024 | Discrete data generation | GitHub |
| Discrete Flow Matching (Gat et al., 2024) | 2024 | Discrete data generation | GitHub |
| Fisher Flow Matching (Davis et al., 2024) | 2024 | Discrete data generation | GitHub |
| Bellman Optimal Stepsize Straightening (Nguyen et al., 2024) | 2024 | Distilling and fine‑tuning flow matching generative models | GitHub |
| Consistency Flow Matching (Yang et al., 2024) | 2024 | Learning strict flows with self‑consistency in velocity fields | GitHub |
| Metric Flow Matching (Kapusniak et al., 2024) | 2024 | Flows on the data manifold | GitHub |
| Depth FM (Gui et al., 2024) | 2024 | Flow matching for depth estimation | GitHub |
| Probabilistic Forecasting with Interpolants (Chen et al., 2024) | 2024 | Stochastic interpolants for probabilistic forecasting of dynamical systems | GitHub |
| SemFlow (Wang et al., 2024) | 2024 | Flows for semantic segmentation & semantic image synthesis | GitHub |
| Preference Flow Matching (Kim et al., 2024) | 2024 | Flow matching for preference‑based reinforcement learning | GitHub |
| COT-FM (Kerrigan et al., 2024) | 2024 | Conditional optimal transport with simulation‑free flows | GitHub |
| FlowSeq (Hu et al., 2024) | 2024 | Flow matching for conditional text generation | GitHub |
| Statistical Flow Matching (Cheng et al., 2024) | 2024 | Flows on the manifold of parameterized probability measures | GitHub |
| Optimal Flow Matching (Kornilov et al., 2024) | 2024 | Shorter flow trajectories | GitHub |
| Discrete Guidance (Nisonoff et al., 2024) | 2024 | Guidance of discrete state‑space flow and diffusion models | GitHub |
| Meta Flow Matching (Atanackovic et al., 2024) | 2024 | Flows across measures and modeling sample interactions | GitHub |
| Trajectory Flow Matching (Zhang et al., 2024) | 2024 | Simulation‑free modeling of stochastic time‑series | GitHub |
| Generator Matching (Holderrieth et al., 2025) | 2024 | Generalized multi‑modal flow matching | GitHub |
| Wasserstein Flow Matching (Haviv et al., 2024) | 2024 | Flows for families of distributions | GitHub |
| Multi-Marginal Flow Matching (Rohbeck et al., 2025) | 2025 | Flows with smooth spline‑based interpolation | GitHub |
| Contrastive Flow Matching (Stoica et al., 2025) | 2025 | Unique conditional flows | GitHub |
| TFG‑Flow (Lin et al., 2025) | 2025 | Continuous‑discrete flow guidance | GitHub |
| SDE Matching (Bartosh et al., 2025) | 2025 | Simulation‑free latent SDEs | GitHub |
| MEND (Saini et al., 2026) | 2026 | Reward post-training of flow models via proximal velocity matching | GitHub |
Application-Specific Flow Matching Models
| Research Area | Applications | Method Name | Flow Conditioning | Network Architecture | Modalities | Data | Performance | Improvement | Source Code |
|---|---|---|---|---|---|---|---|---|---|
| Molecular modeling | Small molecule generation | EquiFM (Song et al., 2023) | Unconditioned | Geometric GNN | Sequence, 3D | GEOM-Drugs | 98.9% (Validity) | +7% | GitHub |
| Small molecule generation | SemlaFlow (Irwin et al., 2024) | Unconditioned | Geometric GNN | Sequence, graph, 3D | GEOM-Drugs | 93.9% (Validity) | -1% | GitHub | |
| Small molecule generation | FlowMol (Dunn et al., 2024) | Unconditioned | Geometric GNN | Sequence, graph, 3D | GEOM-Drugs | 51.2% (Validity) | -33% | GitHub | |
| Small molecule conformer prediction | ET-Flow (Hassan et al., 2024) | Conditioned | Transformer | 3D | GEOM-Drugs | 75.2% (Precision-Coverage) | +13% | GitHub | |
| Small molecule & materials generation | ADiT (Joshi et al., 2025) | Unconditioned | All-atom transformer | Sequence, 3D | GEOM-Drugs | 95.3% (Validity) | +1% | GitHub | |
| Small molecule generation | TABASCO (Vonessen et al., 2025) | Unconditioned | Transformer | Sequence, 3D | GEOM-Drugs | 98.0 (Validity) | +0% | GitHub | |
| Protein sequence generation | ProtFlow (Kong et al., 2025) | Conditioned | Transformer | Sequence | SwissProt | 4.24 (scPerplexity) | +35% | - | |
| Protein structure generation | FrameFlow (Yim et al., 2023) | Unconditioned | SE(3)-equivariant transformer | 3D | SCOPe | 0.81 (Designability) | +93% | GitHub | |
| Protein structure generation | FoldFlow (Bosese & Huguet, 2024) | Unconditioned | SE(3)-equivariant transformer | 3D | PDB | 0.82 (Designability) | +34% | GitHub | |
| Protein side-chain packing | FlowPacker (Lee et al., 2025) | Conditioned | SE(3)-equivariant transformer | 3D | PDB | 0.76Å (RMSD) | +10% | GitHub | |
| Protein sequence & structure generation | Multiflow (Campbell et al., 2024) | Unconditioned | SE(3)-equivariant transformer | Sequence, 3D | PDB | 0.86 (Designability) | +37% | GitHub | |
| Protein sequence & structure generation | CoFlow (Yang et al., 2025) | Conditioned | Transformer | Sequence, 3D | MGnify30 | 3.10Å (scRMSD) | +87% | GitHub | |
| Protein sequence & structure generation | OriginFlow (Yan et al., 2025) | Unconditioned/Conditioned | Transformer | 3D | PDB | 0.45 (TM-Diversity) | +6% | GitHub | |
| Antibody protein sequence & structure | FlowDesign (Wu et al., 2025) | Conditioned | SE(3)-equivariant transformer | Sequence, 3D | SAbDab | 30% (IMP-H3) | +11% | – | |
| Peptide protein sequence & structure | D-Flow (Wu et al., 2024) | Conditioned | Transformer | Sequence, 3D | PDB | 1.63Å (RMSD) | +21% | GitHub | |
| Peptide protein sequence & structure | PepFlow (Li et al., 2024) | Conditioned | SE(3)-equivariant transformer | Sequence, 3D | PDB | 2.07Å (RMSD) | +37% | GitHub | |
| Peptide protein sequence & structure | PPFlow (Lin et al., 2024) | Conditioned | SE(3)-equivariant transformer | Sequence, 3D | PDB | 35.43% (IMP%-B) | +23% | GitHub | |
| DNA sequence generation | Dirichlet FM (Stark et al., 2024) | Unconditioned | Transformer with FM on simplex | Sequence | lncRNA | 1.98 (Perplexity) | +12% | GitHub | |
| DNA sequence generation | Fisher-Flow (Davis et al., 2024) | Unconditioned | Hyperspherical (Fisher-Rao) FM | Sequence | lncRNA | 1.40 (Perplexity) | +29% | GitHub | |
| RNA sequence generation | RNACG (Gao et al., 2024) | Conditioned | Multimodal Diffusion Transformer (Dirichlet FM) | Sequence | PDB | 68.1% (Recall-All) | +2% | – | |
| RNA sequence & structure generation | RNAFlow (Nori et al., 2024) | Conditioned | Geometric GNN | Sequence | PDB | 37.0% (Recovery) | +37% | GitHub | |
| Ligand-conditioned RNA seq & structure | RiboFlow (Ma et al., 2025) | Conditioned | SE(3) flow + torsion modeling | Sequence, 3D | PDB | 34.7% (Validity) | +13% | – | |
| Protein-conditioned RNA seq & structure | RNA-EFM (Abir et al., 2025) | Conditioned | Energy‑based SE(3) flow + protein contacts | Sequence | PDB | 40.0% (Recovery) | +8% | – | |
| RNA structure generation | RNA-FrameFlow (Anand et al., 2024) | Unconditioned | Torsion-based SE(3)-equivariant transformer | 3D | PDB | 25.0% (Validity) | First kind | GitHub | |
| RNA structure prediction | RNAbpFlow (Qiao et al., 2024) | Conditioned | Base-pair-augmented SE(3)-equivariant transformer | 3D | PDB | 0.38 (TM-score) | +19% | GitHub | |
| Biomolecular interactions | FlowDock (Morehead et al., 2025) | Conditioned | SE(3)-equivariant transformer | 3D | PDB | 50.7% (RMSD < 2Å & PB) | +18% | GitHub | |
| Biomolecular interactions | NeuralPLexer3 (Qiao et al., 2024) | Conditioned | All-atom transformer | 3D | PDB | 77.9% (RMSD < 2Å & PB) | +7% | GitHub | |
| Biomolecular interactions | FlexDock (Corso et al., 2025) | Conditioned | SE(3)-equivariant GNN | 3D | PDB | 40.0% (RMSD < 2Å & PB) | +800% | GitHub | |
| Biomolecular interactions | UniMoMo (Kong et al., 2025) | Conditioned | Equivariant transformer | Sequence, graph, 3D | PepBench/ProtFrag, SAbDab, CrossDocked2020 | 10.38 (Weighted Score) | +8% | GitHub | |
| Biomolecular dynamics | AlphaFlow (Jing et al., 2025) | Conditioned | SE(3)-equivariant transformer | 3D | PDB | 0.48 (Pairwise RMSD r) | +218% | GitHub | |
| Biomolecular dynamics | FMRC (Zhang et al., 2024) | Conditioned | Fully-connected neural network | 3D | PDB | 0.45 (Mean eigenvalue=0.38) | +5% | GitHub | |
| Biomolecular dynamics | MDGen (Jing et al., 2024) | Conditioned | Transformer | 3D | MD | 0.48 (Pairwise RMSD r) | +0% | GitHub | |
| Biomolecular dynamics | OM-TPS (Raja et al., 2025) | Conditioned | Diffusion/flow neural network | 3D | MD | 0.35 (JSD) | +40% | GitHub | |
| Biomolecular dynamics | UniSim (Yu et al., 2025) | Conditioned | Geometric GNN | 3D | MD | 0.33 (JSD, pairwise) | +9% | GitHub | |
| Single-cell modeling | Cell phenotype modeling | CellFlow (Klein et al., 2025) | Conditioned | Residual neural network | Sequence | PBMC | 0.60 (R²-DEG) | +200% | GitHub |
| Phenotype generation | CFGen (Palma et al., 2025) | Compositional (multi-attribute) | Flow matching (discrete likelihoods) | Sequence | HLCA | 10.7 (2-WD) | +9% | GitHub | |
| Simulation-free flows | CFM (Tong et al., 2024) | Conditioned | Conditional FM (regression) | Sequence | Multi | 0.94 (1-WD) | +13% | GitHub | |
| Stochastic bridges | [SF]²M (Tong et al., 2023) | Unconditioned | Score‑FM | Sequence | Multi | 0.93 (1-WD) | +1% | GitHub | |
| Stochastic flow alignment | GENOT (Klein et al., 2024) | Unconditioned | Entropic Gromov–Wasserstein flows | Sequence | True Match | 0.16 (FOSCTTM) | +48% | GitHub | |
| Population‑conditioned flow modeling | Meta FM (Atanackovic et al., 2024) | Population‑conditioned | FM‑based GNN | Sequence | PDO | 4.10 (1-WD) | +3% | GitHub | |
| Distribution‑level modeling | Wassertein FM (Haviv et al., 2024) | Distributional | Bures–Wasserstein FM | Sequence | MERFISH | 54.5 (1-WD) | First kind | GitHub | |
| Manifold‑constrained flows | Metric FM (Kapusniak et al., 2024) | Metric‑conditioned | Geodesic interpolant + flow field | Sequence | Multi (100D) | 50.9 (1-WD) | +3% | GitHub | |
| Interaction‑aware bridges | CytoBridge (Zhang et al., 2025) | Conditioned (mean‑field) | Unbalanced mean‑field Schrödinger bridge | Sequence | Mouse | 6.64 (1-WD, t=2) | +2% | – | |
| Rotational dynamics | Curly‑FM (Petrović et al., 2025) | Velocity‑conditioned | Schrödinger bridge with drift | Sequence | Erythroid | 0.49 (Cosine distance, d=50) | +1% | – | |
| Growth‑aware flows | VGFM (Wang et al., 2025) | Conditioned (growth + velocity) | Flow + growth field regression | Sequence | EB | 0.68 (1-WD) | +5% | – | |
| Multi‑context flow learning | MMFM (Rohbeck et al., 2025) | Conditioned (time + condition) | Global vector field + spline guidance | Sequence | Multi (100D) | 70.4 (1-WD) | -38% | GitHub | |
| Multi-cellular modeling | Cell imaging | CellFlux (Zhang et al., 2025) | Conditioned | U‑Net | Image | JUMP | 84.4 (FIDc) | +16% | GitHub |
| Bioimaging | Cryo-EM denoising | CryoFM (Zhou et al., 2025) | Conditioned | Transformer | 3D | EMDB | 0.00 (FR) | +100% | GitHub |
| Microscopic Image Segmentation | FlowSDF (Bogensperger et al., 2025) | Conditioned | U‑Net | Image | MoNuSeg | 81.22 (F1 score) | +0% | GitHub | |
| Biomedical Image Reconstruction | GTF²M (Luo et al., 2025) | Unconditioned | U‑Net | Image | fastMRI | 0.80 (SSIM, 4x) | +1% | - | |
| Biomedical Image Reconstruction | MMSflow (Zhang et al., 2025) | Unconditioned | U‑Net | Image | fastMRI | 0.64 (SSIM, 4x) | +2% | - | |
| Biomedical Image Synthesis | MOTFM (Yazdani et al., 2025) | Conditioned | U‑Net | Image | 3D MSD MRI Brain | 7.93 (3D-FID) | +73% | GitHub |