Google DeepMind·released 2024-07-161 source
Not yet benchmarked
Claims drawn from cited facts, not live model generation.
This model was developed by Google DeepMind and released in 2024, placing it in the current generation of compact open models. Its 2.61 billion parameters make it a compact-scale model, a size that favors efficiency and low deployment cost over the raw headroom of larger counterparts.
3 cited facts
This model is not yet benchmarked, as it is scored on no tracked benchmarks. Because it has no scored benchmarks, it likewise holds no current top score.
2 cited facts
The average gap to state-of-the-art is unknown here, so sota leader count is the deciding reading: this model holds the top score on zero benchmarks, meaning there is no category leadership to point to yet. That is a plain absence of front-of-pack placement rather than a measured distance from the leaders, since no average SOTA gap is disclosed to quantify how far behind it sits.
4 cited facts
2 facts cross-checked across data sources: 2 single-source.
Source facts, citations, and refresh stamp for this record.
Sources: huggingface_models
gemma-2-2b-it is an AI model developed by Google DeepMind, released 2024-07-16. Its benchmark record below tags every score as independently reproduced, self-reported, or unverified.