Google

Local alternatives to Gemini (latest)

Google's current flagship multimodal model.

Google publishes the Gemma open-weight family, which shares Gemini's research lineage and is the closest openly-runnable relative of its hosted Gemini models. Closest by capability class:

These are community-regarded closest open-weight models by capability class — an editorial judgement, not a benchmark ranking. Sizes and hardware/speed figures are derived from the catalog and the same fit engine used across the site, judged at a 8,192-token context. Speeds are estimates, not measurements.

Google open weights

Gemma is Google's own open-weight family, distilled from the same research programme as Gemini — the most defensible "closest local" claim for Gemini.

Local model Size Hardware to run it Est. speed
gemma-4-31B-it google 32.7B Runs fully on 20 GB (RX 7900 XT) at Q4_K_M Runs fully on GPU @ Q4_K_M · 8K ctx 26–35 tok/s (estimate)
gemma-4-26B-A4B-it google tags: Long context 26.5B (MoE) Runs fully on 20 GB (RX 7900 XT) at Q4_K_M Runs fully on GPU @ Q4_K_M · 8K ctx 25–34 tok/s (estimate)
gemma-4-12B-it google 12B Runs fully on 8 GB (RTX 3070) at Q4_K_M Runs fully on GPU @ Q4_K_M · 8K ctx 38–50 tok/s (estimate)

General alternatives

For general-purpose work beyond the Gemma family, the strongest open generalists are the community's usual local choices.

Local model Size Hardware to run it Est. speed
Qwen3-32B Qwen 32.8B Runs fully on 24 GB (RTX 4090) at Q4_K_M Runs fully on GPU @ Q4_K_M · 8K ctx 28–37 tok/s (estimate)

"Hardware to run it" is the smallest single GPU in our roster that runs the model's best-fitting quant fully on GPU at this context, computed by the same fit engine used sitewide. "Est. speed" is a modelled generation-throughput range labelled estimate (D8), never a measurement. Equivalence is a conservative editorial call and no benchmark scores are implied.