Gemma 3 27B
Multimodal
Zero-eval
#1MMMU (val)
#1WMT24++
#1BIG-Bench Extra Hard
+5 more
by Google
+
+
+
+
About
Gemma 3 27B represents the largest variant in the Gemma 3 family, developed to provide flagship-level open-source capabilities. Built with advanced training methodologies and 27 billion parameters, it offers researchers and developers access to powerful language understanding without proprietary restrictions.
+
+
+
+
Pricing Range
Input (per 1M)$0.10 -$0.11
Output (per 1M)$0.20 -$0.20
Providers2
+
+
+
+
Timeline
AnnouncedMar 12, 2025
ReleasedMar 12, 2025
+
+
+
+
Specifications
Training Tokens14.0T
Capabilities
Multimodal
+
+
+
+
License & Family
License
Gemma
Performance Overview
Performance metrics and category breakdown
Overall Performance
26 benchmarks
Average Score
65.4%
Best Score
95.9%
High Performers (80%+)
8Performance Metrics
Max Context Window
262.1KAvg Throughput
33.0 tok/sAvg Latency
0ms+
+
+
+
All Benchmark Results for Gemma 3 27B
Complete list of benchmark scores with detailed information
| GSM8k | text | 0.96 | 95.9% | Self-reported | |
| IFEval | text | 0.90 | 90.4% | Self-reported | |
| MATH | text | 0.89 | 89.0% | Self-reported | |
| HumanEval | text | 0.88 | 87.8% | Self-reported | |
| BIG-Bench Hard | text | 0.88 | 87.6% | Self-reported | |
| DocVQA | multimodal | 0.87 | 86.6% | Self-reported | |
| AI2D | multimodal | 0.84 | 84.5% | Self-reported | |
| Natural2Code | text | 0.84 | 84.5% | Self-reported | |
| ChartQA | multimodal | 0.78 | 78.0% | Self-reported | |
| Global-MMLU-Lite | text | 0.75 | 75.1% | Self-reported |
Showing 1 to 10 of 26 benchmarks
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+