Google model product
95 billion parameters and a 256K context window.
Updated Aug 12, 2026. Default version: Gemma 4 12B
Technical details for the model's default version.
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
Benchmark scores for Gemma 4 12B.
| BIG-Bench Extra Hard | 53.0% | 03 | 11 | 80.0% | C | |
| CoVoST2 | 38.5% | 03 | 4 | 33.3% | C | |
| FLEURS | 93.1% | 04 | 6 | 40.0% | C | |
| MedXpertQA | 48.7% | 08 | 12 | 36.4% | C | |
| MRCR v2 (8-needle) | 43.4% | 09 | 21 | 60.0% | C | |
| MathVision | 79.7% | 13 | 32 | 61.3% | C | |
| AIME 2026 | 77.5% | 15 | 18 | 17.6% | C | |
| CodeForces | 55.3% | 15 | 16 | 6.7% | C | |
| OmniDocBench 1.5 | 16.4% | 15 | 17 | 12.5% | C | |
| LiveCodeBench v6 | 72.0% | 26 | 53 | 51.9% | C | |
| MMMLU | 83.4% | 35 | 49 | 29.2% | C | |
| MMMU-Pro | 69.1% | 38 | 66 | 43.1% | C | |
| MMLU-Pro | 77.2% | 65 | 129 | 50.0% | C | |
| Humanity's Last Exam | 5.2% | 89 | 93 | 4.3% | C | |
| GPQA | 78.8% | 93 | 234 | 60.5% | C |
Preference and agent-evaluation results for the default version.
The default version does not have a matching Arena result yet.
Provider-specific output speed and catalog latency for Gemma 4 12B. Runtime does not affect the capability score.
No provider-specific speed or latency record is linked to the default version yet.
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
Official vendor API pricing appears first, followed by individual provider offers.
| NanoGPT | gemma-4-12b-it | global | $0.06 | $0.30 | 262.1K |
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
Available versions of this model. The score column identifies the version used in the overall ranking.
| Gemma 4 12B | 38.2 | 12B | N/A | N/A | No | Apache 2.0 |
Key information about Gemma 4 12B and its available data.
95 billion parameters and a 256K context window. It supports text, image, audio, and video inputs with text output, projecting image patches and audio waveforms directly into a single decoder-only transformer for streamlined local deployment.
Data as of 2026-08-11.
Open a comparison with the three ranked models immediately above and below this model.
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
Common questions about Gemma 4 12B.
Gemma 4 12B's default version was released on May 23, 2026.
No official standard PAYG price is currently available for Gemma 4 12B. The lowest tracked third-party offer starts at $0.06 input and $0.30 output via NanoGPT.
Gemma 4 12B was created by Google.
A context window is not available for the default version.
No. The default version is not marked as having publicly available weights.
1 provider offerings are linked to the default version.
Nearby ranked alternatives include Mercury 2, Claude Sonnet 3.7, GLM 4.7 Flash.