Gemini 2.5 Pro Preview 06-05
Multimodal
Zero-eval
#1Global-MMLU-Lite
#1FACTS Grounding
#1Vibe-Eval
+5 more
by Google
+
+
+
+
About
Gemini 2.5 Pro Preview 06-05 is a multimodal language model developed by Google. It achieves strong performance with an average score of 68.8% across 13 benchmarks. It excels particularly in Global-MMLU-Lite (89.2%), AIME 2025 (88.0%), FACTS Grounding (87.8%). 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 2025, it represents Google's latest advancement in AI technology.
+
+
+
+
Pricing Range
Input (per 1M)$1.25 -$1.25
Output (per 1M)$10.00 -$10.00
Providers1
+
+
+
+
Timeline
AnnouncedJun 5, 2025
ReleasedJun 5, 2025
Knowledge CutoffJan 31, 2025
+
+
+
+
Specifications
Capabilities
Multimodal
+
+
+
+
License & Family
License
Proprietary
Performance Overview
Performance metrics and category breakdown
Overall Performance
13 benchmarks
Average Score
68.8%
Best Score
89.2%
High Performers (80%+)
7Performance Metrics
Max Context Window
1.1MAvg Throughput
85.0 tok/sAvg Latency
1ms+
+
+
+
All Benchmark Results for Gemini 2.5 Pro Preview 06-05
Complete list of benchmark scores with detailed information
| Global-MMLU-Lite | text | 0.89 | 89.2% | Self-reported | |
| AIME 2025 | text | 0.88 | 88.0% | Self-reported | |
| FACTS Grounding | text | 0.88 | 87.8% | Self-reported | |
| GPQA | text | 0.86 | 86.4% | Self-reported | |
| VideoMMMU | multimodal | 0.84 | 83.6% | Self-reported | |
| Aider-Polyglot | text | 0.82 | 82.2% | Self-reported | |
| MMMU | multimodal | 0.82 | 82.0% | Self-reported | |
| LiveCodeBench | text | 0.69 | 69.0% | Self-reported | |
| SWE-Bench Verified | text | 0.67 | 67.2% | Self-reported | |
| Vibe-Eval | multimodal | 0.67 | 67.2% | Self-reported |
Showing 1 to 10 of 13 benchmarks