Llama-3.2-1B-Instruct (unmodified redistribution)
Built with Llama.
This is an unmodified redistribution of
meta-llama/Llama-3.2-1B-Instruct at upstream commit
e9f8effbab1cbdc515c11ee6e098e3d5a9f51e14.
Weights, tokenizer, and configuration files are byte-identical to that commit
(SHA256s below); only this README and the NOTICE file were added.
Positron AI redistributes this snapshot as the pinned source for the GPTQ
quantizations it publishes under the positron-ai organization.
Model documentation, evaluations, and responsible-use guidance: see Meta's original model card linked above.
Provenance
| file | sha256 |
|---|---|
| model.safetensors | 1ff795ff6a07e6a68085d206fb84417da2f083f68391c2843cd2b8ac6df8538f |
License
Use of this model is governed by the
Llama 3.2 Community License Agreement and Meta's
Acceptable Use Policy, both included in this repository.
See NOTICE. All model documentation, benchmarks, and responsible-use guidance
are at the upstream model card linked above.