Atom-7B safetensors download | Hugging Bay

Q: What is this model?

FlagAlpha/Atom-7B is an open-weights language model for question answering published by FlagAlpha, added to Hugging Bay Sep 24, 2026. Model from Hugging Face for question answering: FlagAlpha/Atom-7B · transformers

Q: Can I use it commercially?

It is released under the apache-2.0 license. Commercial use is allowed. Permissive open license: commercial use, modification, and redistribution are permitted.

Q: How big is it?

FlagAlpha/Atom-7B has 7B parameters. Hosted here: 9 files, 14 GB. Full repo upstream: 14 GB. The recommended download, a 3-file weight set, is 14 GB.

Q: What can I download here?

Hosted here: 9 files, 14 GB. Full repo upstream: 14 GB. The original release is at https://huggingface.co/FlagAlpha/Atom-7B.

Q: File pattern scans

Automated file-pattern scans flagged nothing; not a safety guarantee. 9 hosted files carry a published SHA-256 you can re-hash. The machine-readable verdict is at /api/artifacts/hf-model-flagalpha-atom-7b/safety on Hugging Bay.

Q: Will FlagAlpha/Atom-7B run on my computer?

We can’t check the fit; we don’t know your machine. We don’t have your machine’s specs, so no check ran and no verdict is shown. Answer two quick questions and we’ll estimate, from the memory you tell us, whether it fits your machine. Or skip it; you can still download; fit isn’t checked.

FlagAlpha

Atom-7B

FlagAlpha/Atom-7B

Model details

question answeringLicense: apache-2.0SHA-256 recorded for 9 of 14 filessafetensors

Content controls: not assessed.

Listed for answering questions; for developers building question answering tools. Catalog sources: license, cardMetadata.license, task, cardMetadata.pipelineTag.

View as JSON →

9 hosted of 14 listed

Hashes recorded. Date not recorded.

File	Taille	SHA-256	Source	Télécharger
Weights
`model-00001-of-00003.safetensors`	4.98 GB	`4cb46b55…a665bbb7` `4cb46b554509faa3864b220c7cca8093e8c7cc3e57efd3f25391ff86a665bbb7`	Hosted	Télécharger
`model-00002-of-00003.safetensors`	4.99 GB	`0518a003…5a600ab2` `0518a0031d46b170e17d8c3d354b8e91a17c536cfbb807b0c3cb096c5a600ab2`	Hosted	Télécharger
`model-00003-of-00003.safetensors`	4.04 GB	`214059ad…7c44e1eb` `214059ad3834a892d600c97aeba063e7f87d6a439e3b0b97e8e7a82b7c44e1eb`	Hosted	Télécharger
Tokenizer & config
`config.json`	1.07 KB	`061ed0f8…e0ed9c25` `061ed0f8dec5cad3e1357ede455ada8edf7cbe38e6edea0f979fe435e0ed9c25`	Hosted	Télécharger
`configuration_atom.py`	9.87 KB	—	Upstream only	File metadata
`generation_config.json`	154 bytes	`1a3d5986…95434df1` `1a3d59866ef3f0f0bf2d36012efed2b8e80166d2596feb1379b9ecf095434df1`	Hosted	Télécharger
`model.safetensors.index.json`	23.9 KB	`2a96b2c0…5ea7ee06` `2a96b2c090dc31c1c8340e3094694244474e40a22a7df1224e3ac12d5ea7ee06`	Hosted	Télécharger
`model_atom.py`	73.2 KB	—	Upstream only	File metadata
`special_tokens_map.json`	548 bytes	`a427b7e0…d3f74829` `a427b7e082f9f34ddd5a9a224c6ddcb60d960b2dcbc859f0f05a67bfd3f74829`	Hosted	Télécharger
`tokenizer.model`	1.01 MB	`04ef61cc…7e7bcdd8` `04ef61cc08360cd193f9056cb10e26525451fd62759ca714840663257e7bcdd8`	Hosted	Télécharger
`tokenizer_config.json`	932 bytes	`43108daf…04769423` `43108daffbec6cfad85c6a62e82ecb3db3a8c00021262623ad54fe4504769423`	Hosted	Télécharger
`zero_to_fp32.py`	24.2 KB	—	Upstream only	File metadata
Docs
`.gitattributes`	1.52 KB	—	Upstream only	File metadata
`README.md`	5.1 KB	—	Upstream only	File metadata

Vérifier un téléchargement

In the folder where you saved config.json, run:

macOS

shasum -a 256 -- 'config.json'

Linux

sha256sum -- 'config.json'

Windows

Get-FileHash -LiteralPath 'config.json' -Algorithm SHA256

Compare the result with the recorded SHA-256.

Modèles hébergés similaires

Explorer les modèles hébergés

Atom-7B

Vérifier un téléchargement

Modèles hébergés similaires

Cookies