TranslateGemma 12B
TranslateGemma 12B, developed by Google DeepMind, is a foundation model specialized for translation. It is equipped with a long context window of 128K and supports advanced translation tasks.
Parameters
13.0B
Context Window
128K
License
Gemma License
Release Date
2026-01-15
API Pricing
API pricing for this model is not yet available
Strengths
- ・High specialization in translation
- ・128K broad context window
- ・Developed by Google DeepMind
Weaknesses
- ・Unknown adaptability to general tasks
- ・Load from ~25GB model size
- ・Subject to usage terms restrictions
Use Cases
- ・High-precision translation of long documents
- ・Localization of multilingual content
- ・Building specialized translation pipelines
Deep Analysis
Parameters
12B
Mid-size TranslateGemma variant
Performance
Outperforms Gemma 3 27B baseline
Uses less than half the parameters to exceed 27B baseline quality
Languages
55 core / ~500 extended
Comprehensive language coverage
Benchmark
WMT24++ MetricX
Tested on 55-language WMT24++ dataset
Release Date
January 15, 2026
Part of TranslateGemma family launch
Multimodal
Yes
Image text translation via Vistra benchmark
Training
SFT + RL
Two-stage: supervised fine-tuning + reinforcement learning with ensemble reward models
Strengths
- ・Outperforms Gemma 3 27B baseline on translation with less than half the parameters
- ・Best balance of quality and efficiency in the TranslateGemma family
- ・55 core languages with multimodal image translation
- ・Distilled from Gemini models for high fidelity
- ・Open weights, available on Kaggle and Hugging Face
- ・Suitable for server-side deployment with good throughput
Weaknesses
- ・Lower quality than 27B variant for the most demanding translation tasks
- ・Not a general-purpose model (translation-focused)
- ・Extended language pairs lack confirmed metrics
- ・Larger than 4B, not suitable for mobile
- ・Limited community benchmarking outside Google's evaluations
Competitor Comparison
| Model | Arena | SWE | GPQA | Price |
|---|---|---|---|---|
| TranslateGemma 12B | N/A | N/A | N/A | Free (open weights) |
| TranslateGemma 27B | N/A | N/A | N/A | Free (open weights) |
| Gemma 3 27B (baseline) | ~1430 | N/A | ~70% | Free (open weights) |
| NLLB-200 (54.5B) | N/A | N/A | N/A | Free (open weights) |
| GPT-5 (general) | ~1480 | ~80% | ~90% | $5/$20 per 1M |
TranslateGemma 12B is the mid-size variant that outperforms the Gemma 3 27B baseline on translation quality using less than half the parameters. It represents the best efficiency breakthrough in the TranslateGemma family, achieving high-fidelity translation across 55 languages.
Sources
Analysis generated: 2026-05-24