Small fast 'System 1' · Meta
Llama 3.2 1B
Meta's 1B model for on-device summarising and rewriting.
Datasheet
Maker | Meta |
|---|---|
Parameters | 1.24 BVerifiedsafetensors.total from the HF API Source |
Licence | Llama 3.2 CommunityVerified |
Access | Gated: accept the terms on Hugging Face to download |
Task (HF pipeline) | text-generation |
First published on HF | 2024-09-18 |
Last updated on HF | 2024-10-24 |
Weights in memory | 2.5 GB at 16-bit · 0.7 GB at ~4.85 bitsEstimateParameters × bits ÷ 8. Excludes context cache and runtime overhead; not every model has a 4-bit build. |
Benchmark
10.82%
BFCL V4, overall accuracy · #2 of 3 scored in this category
Llama-3.2-1B-Instruct (FC) row, rank 107. Source
The description summarises the model card at https://huggingface.co/meta-llama/Llama-3.2-1B-Instruct. Parameters and licence are read from the Hugging Face API; the weights-in-memory line is our arithmetic, not a measurement.