Gemini 1.5 Flash
Multimodal
Zero-eval
#2XSTest
#2WMT23
#2PhysicsFinals
+5 more
by Google
+
+
+
+
About
Gemini 1.5 Flash is a multimodal language model developed by Google. It achieves strong performance with an average score of 66.8% across 22 benchmarks. It excels particularly in XSTest (97.0%), HellaSwag (86.5%), GSM8k (86.2%). With a 1.1M token context window, it can handle extensive documents and complex multi-turn conversations. 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 2024, it represents Google's latest advancement in AI technology.
+
+
+
+
Pricing Range
Input (per 1M)$0.15 -$0.15
Output (per 1M)$0.60 -$0.60
Providers1
+
+
+
+
Timeline
AnnouncedMay 1, 2024
ReleasedMay 1, 2024
Knowledge CutoffNov 1, 2023
+
+
+
+
Specifications
Capabilities
Multimodal
+
+
+
+
License & Family
License
Proprietary
Performance Overview
Performance metrics and category breakdown
Overall Performance
22 benchmarks
Average Score
66.8%
Best Score
97.0%
High Performers (80%+)
5Performance Metrics
Max Context Window
1.1MAvg Throughput
150.0 tok/sAvg Latency
0ms+
+
+
+
All Benchmark Results for Gemini 1.5 Flash
Complete list of benchmark scores with detailed information
| XSTest | text | 0.97 | 97.0% | Self-reported | |
| HellaSwag | text | 0.86 | 86.5% | Self-reported | |
| GSM8k | text | 0.86 | 86.2% | Self-reported | |
| BIG-Bench Hard | text | 0.85 | 85.5% | Self-reported | |
| MGSM | text | 0.83 | 82.6% | Self-reported | |
| Natural2Code | text | 0.80 | 79.8% | Self-reported | |
| MMLU | text | 0.79 | 78.9% | Self-reported | |
| MATH | text | 0.78 | 77.9% | Self-reported | |
| Video-MME | multimodal | 0.76 | 76.1% | Self-reported | |
| HumanEval | text | 0.74 | 74.3% | Self-reported |
Showing 1 to 10 of 22 benchmarks