Gemma 3n E4B Instructed
Multimodal
Zero-eval
#1Global-MMLU
#1OpenAI MMLU
#1ECLeKTic
+1 more
by Google
+
+
+
+
About
Gemma 3n E4B Instructed is a multimodal language model developed by Google. The model shows competitive results across 18 benchmarks. Notable strengths include HumanEval (75.0%), MGSM (67.0%), MMLU (64.9%). The model is available through 1 API provider. As a multimodal model, it can process and understand text, images, and other input formats seamlessly. Released in 2025, it represents Google's latest advancement in AI technology.
+
+
+
+
Pricing Range
Input (per 1M)$20.00 -$20.00
Output (per 1M)$40.00 -$40.00
Providers1
+
+
+
+
Timeline
AnnouncedJun 26, 2025
ReleasedJun 26, 2025
Knowledge CutoffJun 1, 2024
+
+
+
+
Specifications
Training Tokens11.0T
Capabilities
Multimodal
+
+
+
+
License & Family
License
Proprietary
Performance Overview
Performance metrics and category breakdown
Overall Performance
18 benchmarks
Average Score
42.0%
Best Score
75.0%
High Performers (80%+)
0Performance Metrics
Max Context Window
64.0KAvg Throughput
42.1 tok/sAvg Latency
0ms+
+
+
+
All Benchmark Results for Gemma 3n E4B Instructed
Complete list of benchmark scores with detailed information
| HumanEval | text | 0.75 | 75.0% | Self-reported | |
| MGSM | text | 0.67 | 67.0% | Self-reported | |
| MMLU | text | 0.65 | 64.9% | Self-reported | |
| Global-MMLU-Lite | text | 0.65 | 64.5% | Self-reported | |
| MBPP | text | 0.64 | 63.6% | Self-reported | |
| Global-MMLU | text | 0.60 | 60.3% | Self-reported | |
| Include | text | 0.57 | 57.2% | Self-reported | |
| MMLU-Pro | text | 0.51 | 50.6% | Self-reported | |
| WMT24++ | text | 0.50 | 50.1% | Self-reported | |
| HiddenMath | text | 0.38 | 37.7% | Self-reported |
Showing 1 to 10 of 18 benchmarks