Meta AI·released 2024-09-181 source
Not yet benchmarked
Claims drawn from cited facts, not live model generation.
Developed by Meta AI and released in 2024, this model is a compact 1.24-billion-parameter model whose small scale favors efficiency and lower cost. That compact footprint trades raw capacity for affordability, making it suited to constrained deployments rather than frontier-scale workloads.
4 cited facts
No benchmarks are tracked for this model, so it is not yet benchmarked and holds no current top score.
2 cited facts
Average SOTA gap is unknown, so there is no points-behind figure to place on the frontier-to-behind band; instead the record shows it holds the top score on 0 benchmarks. Plainly, that means no category leadership yet — it is not front-of-pack on any of its benchmarks.
3 cited facts
6 facts cross-checked across data sources: 6 single-source.
Source facts, citations, and refresh stamp for this record.
Sources: huggingface_models, openrouter_models
Llama-3.2-1B-Instruct is an AI model developed by Meta AI, released 2024-09-18. Its benchmark record below tags every score as independently reproduced, self-reported, or unverified.