Find open-source science resources
A directory of tools, AI models, datasets, and research resources for biotech, bioinformatics, and other scientific fields. Aggregated from curated GitHub awesome-lists, HuggingFace, bio.tools, Bioconductor, and more.
Filters
Health
Domain
Language
License
Source
Type(1)
1,095 of 9,874 resources
Showing 51–100
CENO-300M-A105M-base is the base pretraining checkpoint of the 300M CENO DNA foundation model — a causal language model over genomic sequence built on a Nemotron-H Mamba / Attention / Mixture-of-Experts hybrid backbone (no MSA inputs).
CladeTeam/CENO-80M-A30M-1m
by CladeTeamCENO-80M-A30M-1m is the long-context (1M) checkpoint of the 80M CENO DNA foundation model — a causal language model over genomic sequence built on a Nemotron-H Mamba / Attention / Mixture-of-Experts hybrid backbone (no MSA inputs).
CladeTeam/CENO-80M-A30M-131k
by CladeTeamCENO-80M-A30M-131k is the long-context (131k) checkpoint of the 80M CENO DNA foundation model — a causal language model over genomic sequence built on a Nemotron-H Mamba / Attention / Mixture-of-Experts hybrid backbone (no MSA inputs).
CladeTeam/CENO-80M-A30M-base
by CladeTeamCENO-80M-A30M-base is the base pretraining checkpoint of the 80M CENO DNA foundation model — a causal language model over genomic sequence built on a Nemotron-H Mamba / Attention / Mixture-of-Experts hybrid backbone (no MSA inputs).
CladeTeam/CENO-P-1B
by CladeTeamCENO-P-1B is the multi-species alignment (MSA) post-trained variant of the 1B CENO DNA foundation model, for variant effect prediction (VEP). It carries intraencodingpattern in its config and ships the MSA scoring path (modelingcenop.py), which consumes a per-token seq_idx to score packed MSA…
CladeTeam/CENO-P-600M
by CladeTeamCENO-P-600M is the multi-species alignment (MSA) post-trained variant of the 600M CENO DNA foundation model, for variant effect prediction (VEP). It carries intraencodingpattern in its config and ships the MSA scoring path (modelingcenop.py), which consumes a per-token seq_idx to score packed MSA…
CladeTeam/CENO-P-300M
by CladeTeamCENO-P-300M is the multi-species alignment (MSA) post-trained variant of the 300M CENO DNA foundation model, for variant effect prediction (VEP). It carries intraencodingpattern in its config and ships the MSA scoring path (modelingcenop.py), which consumes a per-token seq_idx to score packed MSA…
HuggingFaceBio/Carbon-A-1.2B
by HuggingFaceBioA DNA annotation model from the Carbon family.
pavanmantha/mednli_laya
by pavanmanthaLaya decision model fine-tuned as a three-way medical claim checker. Given a claim and a piece of evidence, it answers supported, unsupported or contradicted.
gevaertlab/MANTA-4B
by gevaertlabMANTA (Medical Answers, Natively from Text And images) is a medical decision model developed by the Gevaert Lab at the Department of Computational Medicine, Stanford University. Give it a medical image or text and a set of questions with their answer options: MANTA-4B returns a probability for…
gevaertlab/MANTA-27B
by gevaertlabMANTA (Medical Answers, Natively from Text And images) is a medical decision model developed by the Gevaert Lab at the Department of Computational Medicine, Stanford University. Give it a medical image or text and a set of questions with their answer options: MANTA-27B returns a probability for…
OpenMed Qwen3-1.7B WSP Clinical
> DON'T PANIC. Welcome to Deep Thought, ostensibly the second greatest computational matrix in the universe. While its predecessor spent 7.5 million years just to output "42," this modern iteration is heavily fine-tuned on the galaxy’s most rigorous STEM and hard science datasets, powered by a…
> DON'T PANIC. Welcome to Deep Thought, ostensibly the second greatest computational matrix in the universe. While its predecessor spent 7.5 million years just to output "42," this modern iteration is heavily fine-tuned on the galaxy’s most rigorous STEM and hard science datasets, powered by a…
appliedCL/interseek-35m
by appliedCLInterSeek predicts whether two proteins interact from their amino-acid sequences. This release contains the complete pair predictor and the fine-tuned ESM-2 backbone from a completed, single-seed experiment.
joelinator/dflow-novo-model
by joelinator## Task Description De novo peptide sequencing from tandem mass spectrometry (MS/MS) using discrete flow matching over Continuous-Time Markov Chains (CTMC).
biohub/ESMC-6B
by biohubESMC is a state-of-the-art protein language model that has learned the rules of protein biology from training on billions of protein sequences. ESMC provides representations of proteins enabling novel AI applications from therapeutic protein engineering to unlocking basic insights into protein…
internlm/Sage-SciPerception
by internlmSage-SciPerception Scientific perception for language models · 书生·元征 Connect crystals, spectra, biological sequences, and biosignals to language models through trained projectors. Continuously expanding scientific modalities, tasks, and language model support.
CHM4930/mufu
by CHM4930MuFu, from CHM4930, fuses a frozen MoLFormer SMILES embedding, an ECFP4 fingerprint and a GINE graph encoder. For one SMILES it predicts EGFR and BACE1 pIC50 (with a conformal interval) and P(active), the 12 Tox21 assays and the 2 ClinTox tasks.
atom-jepa/atom-jepa
by atom-jepaPretrained encoders from Atom-JEPA: Joint-Embedding Predictive Architecture for 3D Atomistic Systems.
plantcad/marin-dna-exp586-1.5B-60M
by plantcadBF16 export of the final 1T-token exp586 checkpoint: 1.59B total parameters and approximately 60.6M active parameters per token.
Scinetics-ai/all-atom-structure-tokenizer
by Scinetics-aiA PyTorch reference implementation of the tokenizer described in the Scinetics AtomLM-9b for Biomolecular Structure Prediction technical report. The encoder converts a heavy-atom assembly to one integer per chemical unit. The coordinate decoder reconstructs an assembly from those integers, its…
STORMResearch/locus
by STORMResearchPretrained LoCUS models: DNA language models that read DNA as segments from locally consistent parsing (LCP). Code, fine-tuning recipes and the scripts that reproduce every result of the paper: https://github.com/STORMgroup/locus
InstaDeepAI/NTv3_pretrained
by InstaDeepAIziyux123/chefnmr2
by ziyux123Weights of ChefNMR2, which elucidates the structure of a small molecule from its 1H and 13C NMR peak lists and molecular formula.
gaozijun/CellPlex
by gaozijunRConci/GENESIS
by RConciendless-frontier/Fx-Bio-1031
by endless-frontierFx-Bio-1031
mradermacher/BioMistral-Clinical-7B-GGUF
by mradermacherFor a convenient overview and download list, visit our model page for this model.
maltefranke/morph
by maltefrankeCheckpoints for unconditional 3D molecular generation on QM9, GEOM-Drugs and ChEMBL3D, GEOM-Drugs N-conditional generation, GEOM-Drugs and QM9 logP-conditional generation, GEOM-Drugs scaffold decoration, and CrossDocked scaffold decoration in protein pockets.
yangbai123/EVLF-FM
by yangbai123EVLF-FM is a medical multimodal model for disease diagnosis, visual grounding, and visual question answering. It is based on InternVL3-8B. This version is trained on publicly available datasets.
arcinstitute/bolero-models
by arcinstituteTrained checkpoints for Bolero, a sequence-to-function model that predicts cell-state-specific chromatin accessibility and gene expression from DNA sequence. Each checkpoint holds LoRA weights trained on a Borzoi backbone.
programmable-genomics/BATv1
by programmable-genomicsThis is the first release of the Cherimoya Binding ATlas (BATv1): a collection of 17,145 Cherimoya models trained on ChIP-seq experiments from the ENCODE Project that measure the binding of transcription factors and other DNA-associated proteins.
OpenMed GLiFormer BioNER Disease Base v1
OpenMed GLiNER2.5 BioNER Disease Base v1
HuggingFaceBio/Carbon-X
by HuggingFaceBioBlog | Hugging Face Bio | Code
HuggingFaceBio/Carbon-3B
by HuggingFaceBioTechnical Report 🧬
All encoder, learned atomic/geometric preprocessor, and COSMO-head weights were initialized from the exact PCQM epoch 2 → COSMO epoch 10 checkpoint used in the preceding BigSolDB 2.0 evaluation. Byte-level hashes were checked against that evaluation protocol.
sachin-kukkar/ai-nose-openpom
by sachin-kukkar10 message-passing neural networks (OpenPOM MPNNPOMModel) that predict 138 odor descriptors from a molecule's structure, plus a 256-dimensional "odor map" embedding for every catalogue molecule.
bios-labs/RFD4-Proteina-MLX-fp16
by bios-labsDesign protein binders with RFD4-Proteina entirely on a MacBook: target structure in, ranked and co-fold-screened binders out. No CUDA and no cloud. After the one-time downloads, this repo's commands need no network (set HFHUBOFFLINE=1 to enforce it for Hugging Face lookups; python -m rfd4_mlx.prep…
InstaDeepAI/NTv3_650M_pre
by InstaDeepAIQueimo/rem3di-pcqm-epoch2-cosmo
by QueimoInitialization: Queimo/rem3di-pcqm500k-corrected at commit 2f5c3b40b36259a13adc8412689254054bb93288, runs/corrected20261027/evaluation/paperaudit/authorpcqm500k5ep/epoch_02.
westlake-repl/Pinal
by westlake-replPinal is an advanced protein design framework that translates human design intent into novel protein sequences. It utilizes a two-stage process: first generating protein structures from language instructions, then designing sequences based on those structures.
HuggingFaceBio/Carbon-X-SFT
by HuggingFaceBioBlog | Carbon-X | Carbon-X-Base | Hugging Face Bio | Code