{"success":true,"data":{"author":"Arc Institute","bucketAccessType":"public","bucketName":"models/arcinstitute/evo2_7b","bucketPrefix":"","category":"Genomics & Bioinformatics","cloudProvider":"huggingface","createdAt":"2026-11-16T15:25:49.168Z","description":"Open weights for Evo 2 7B, Arc Institute's DNA language model trained autoregressively on trillions of nucleotide tokens from the OpenGenome2 corpus spanning all domains of life, with a context of up to one million base pairs. The 32-layer StripedHyena 2 model scores variant effects, embeds genomic sequence and generates DNA. The mounted repo holds evo2_7b.pt (13.8 GB) and config.json; load it with the evo2 package by pointing the checkpoint path at the mount, and see the Evo 2 repository for its GPU requirements. The OpenGenome2 training corpus is also in this catalog. Apache 2.0.","displayTitle":"Evo 2 7B DNA language model weights (1M context)","huggingFaceUrl":"https://huggingface.co/arcinstitute/evo2_7b","id":"e5b24ac6-c3ad-5a3f-a84f-7a73b2539a82","license":"Apache-2.0","licenseEvidenceUrl":"https://github.com/ArcInstitute/evo2/blob/main/LICENSE","objectCountEstimate":4,"probeStatus":"ok","referenceUrl":"https://www.biorxiv.org/content/10.1101/2025.02.18.638918v1","region":"us","schemaCard":{"accessHints":[{"language":"python-hub","snippet":"from huggingface_hub import snapshot_download\nsnapshot_download(repo_id=\"arcinstitute/evo2_7b\")  # or read /mnt/archive directly"}],"caveats":["HuggingFace metadata: author=arcinstitute, license:apache-2.0, downloads=1804","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":"2026-11-17T19:17:23.610Z","layout":{"example":"evo2_7b.pt","pattern":"{repoId}/{filename}"},"mountPath":"/mnt/archive","sampleObjects":["config.json","evo2_7b.pt"],"sidecars":[{"bytes":"---\nlicense: apache-2.0\ndatasets:\n- arcinstitute/opengenome2\n---\n\n<img src=\"https://cdn-uploads.huggingface.co/production/uploads/649aee789fc303937a045f6a/IGUfG31MMvDzhdjRK-nlJ.jpeg\" width=\"70%\" />\n\n## Evo 2\n\nEvo 2 is a state-of-the-art DNA language model trained autoregressively on trillions of DNA tokens.\n\nFor instructions, details, and examples, please refer to the [Evo 2 gitHub](https://github.com/ArcInstitute/evo2) and [paper](https://www.biorxiv.org/content/10.1101/2025.02.18.638918v1).\n\nEvo 2 40B and 7B checkpoints, trained up to 1 million sequence length, are available here:\n| Checkpoint name | Num layers | Num parameters |\n|------------------------------|----|----------|\n| [evo2_40b](https://huggingface.co/arcinstitute/evo2_40b) | 50 | 40B |\n| [evo2_7b](https://huggingface.co/arcinstitute/evo2_7b) | 32 | 7B  | \n\nWe also share 40B, 7B, and 1B base checkpoints trained on 8192 context length:\n| Checkpoint name | Num layers | Num parameters |\n|------------------------------|----|----------|\n| [evo2_40b_base](https://huggingface.co/arcinstitute/evo2_40b_base) | 50 | 40B |\n| [evo2_7b_base](https://huggingface.co/arcinstitute/evo2_7b_base) | 32 | 7B  | \n| [evo2_1b_base](https://huggingface.co/arcinstitute/evo2_1b_base) | 25 | 1B  |","encoding":"utf8","key":"README.md","name":"README.md","size":1253,"truncated":false}],"sourceObjectKey":"evo2_7b.pt","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])\nprint(\"evo2_7b.pt\", os.path.getsize(os.path.join(base, \"evo2_7b.pt\")) // 2**20, \"MB\")","totalObjects":4},"summary":"Open weights for Evo 2 7B, Arc Institute's DNA language model trained autoregressively on trillions of nucleotide tokens from the OpenGenome2 corpus spanning all domains of life,…","title":"Evo 2 7B DNA language model weights (1M context)","updatedAt":"2026-11-17T19:17:23.893Z","visibility":"public","websiteUrl":"https://huggingface.co/arcinstitute/evo2_7b"}}