{"success":true,"data":{"author":"Meta AI (FAIR)","bucketAccessType":"public","bucketName":"models/facebook/esm2_t33_650M_UR50D","bucketPrefix":"","category":"Biology","cloudProvider":"huggingface","createdAt":"2026-11-16T15:25:49.027Z","description":"Open weights for ESM-2 650M, the 33-layer, 650 million parameter protein language model from Meta's ESM project (Lin et al., Science 2023), trained with masked language modelling on UniRef50 protein sequences. It produces per-residue and per-sequence embeddings used for fitness prediction, contact prediction, function annotation and as the base model for many fine-tuned protein predictors. The mounted repo holds model.safetensors, pytorch_model.bin and tf_model.h5 (about 2.6 GB each), config.json, vocab.txt and tokenizer files, and loads with transformers (EsmModel, EsmForMaskedLM) from the mount path. Pair it with the UniRef, ProteinGym or PDB archives to embed sequences next to the data. MIT licence.","displayTitle":"ESM-2 650M protein language model weights (esm2_t33_650M_UR50D)","huggingFaceUrl":"https://huggingface.co/facebook/esm2_t33_650M_UR50D","id":"5c5569aa-65ba-5e25-9da3-d07fb16c9664","license":"MIT","licenseEvidenceUrl":"https://github.com/facebookresearch/esm/blob/main/LICENSE","objectCountEstimate":9,"probeStatus":"ok","referenceUrl":"https://www.science.org/doi/10.1126/science.ade2574","region":"us","schemaCard":{"accessHints":[{"language":"python-hub","snippet":"from huggingface_hub import snapshot_download\nsnapshot_download(repo_id=\"facebook/esm2_t33_650M_UR50D\")  # or read /mnt/archive directly"}],"caveats":["HuggingFace metadata: author=facebook, license:mit, library=transformers, pipeline=fill-mask, downloads=3263441","Weights are read lazily from the mount. Copying them to local disk first is faster when a model is loaded many times."],"format":"binary","generatedAt":"2026-11-17T19:17:19.336Z","layout":{"example":"model.safetensors","pattern":"{repoId}/{filename}"},"mountPath":"/mnt/archive","sampleObjects":["config.json","model.safetensors","pytorch_model.bin","tf_model.h5"],"sidecars":[{"bytes":"---\nlicense: mit\nwidget:\n - text: \"MQIFVKTLTGKTITLEVEPS<mask>TIENVKAKIQDKEGIPPDQQRLIFAGKQLEDGRTLSDYNIQKESTLHLVLRLRGG\"\n---\n\n## ESM-2\n\nESM-2 is a state-of-the-art protein model trained on a masked language modelling objective. It is suitable for fine-tuning on a wide range of tasks that take protein sequences as input. For detailed information on the model architecture and training data, please refer to the [accompanying paper](https://www.biorxiv.org/content/10.1101/2022.07.20.500902v2). You may also be interested in some demo notebooks ([PyTorch](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/protein_language_modeling.ipynb), [TensorFlow](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/protein_language_modeling-tf.ipynb)) which demonstrate how to fine-tune ESM-2 models on your tasks of interest.\n\nSeveral ESM-2 checkpoints are available in the Hub with varying sizes. Larger sizes generally have somewhat better accuracy, but require much more memory and time to train:\n\n| Checkpoint name | Num layers | Num parameters |\n|------------------------------|----|----------|\n| [esm2_t48_15B_UR50D](https://huggingface.co/facebook/esm2_t48_15B_UR50D) | 48 | 15B     |\n| [esm2_t36_3B_UR50D](https://huggingface.co/facebook/esm2_t36_3B_UR50D) | 36 | 3B      | \n| [esm2_t33_650M_UR50D](https://huggingface.co/facebook/esm2_t33_650M_UR50D) | 33 | 650M    | \n| [esm2_t30_150M_UR50D](https://huggingface.co/facebook/esm2_t30_150M_UR50D) | 30 | 150M    | \n| [esm2_t12_35M_UR50D](https://huggingface.co/facebook/esm2_t12_35M_UR50D) | 12 | 35M     | \n| [esm2_t6_8M_UR50D](https://huggingface.co/facebook/esm2_t6_8M_UR50D)  | 6  | 8M      | ","encoding":"utf8","key":"README.md","name":"README.md","size":1705,"truncated":false}],"sourceObjectKey":"model.safetensors","starterCell":"import os, json\nbase = \"/mnt/archive\"   # this model repo, mounted read-only\nfiles = sorted(os.listdir(base)); print(files)\nprint(json.dumps(json.load(open(os.path.join(base, \"config.json\"))), indent=1)[:800])\nimport struct\nwith open(os.path.join(base, \"model.safetensors\"), \"rb\") as f:\n    hdr = json.loads(f.read(struct.unpack(\"<Q\", f.read(8))[0]))\ntensors = {k: v for k, v in hdr.items() if k != \"__metadata__\"}\nprint(len(tensors), \"tensors, e.g.\", list(tensors.items())[:3])\n# To load them: pip install safetensors torch, then safetensors.torch.load_file(path)","totalObjects":9},"summary":"Open weights for ESM-2 650M, the 33-layer, 650 million parameter protein language model from Meta's ESM project (Lin et al., Science 2023), trained with masked language modelling…","title":"ESM-2 650M protein language model weights (esm2_t33_650M_UR50D)","updatedAt":"2026-11-17T19:17:19.608Z","visibility":"public","websiteUrl":"https://huggingface.co/facebook/esm2_t33_650M_UR50D"}}