Google model product
5 Pro is a mid-size multimodal model optimized for a wide range of reasoning tasks.
Updated Aug 12, 2026. Default version: Gemini 1.5 Pro
Technical details for the model's default version.
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
Benchmark scores for Gemini 1.5 Pro.
| FunctionalMATH | 64.6% | 01 | 2 | 100.0% | C | |
| PhysicsFinals | 63.9% | 01 | 2 | 100.0% | C | |
| WMT23 | 75.1% | 01 | 4 | 100.0% | C | |
| XSTest | 98.8% | 01 | 3 | 100.0% | C | |
| MRCR | 82.6% | 02 | 7 | 83.3% | C | |
| Natural2Code | 85.4% | 02 | 8 | 85.7% | C | |
| BIG-Bench Hard | 89.2% | 03 | 21 | 90.0% | C | |
| FLEURS | 93.3% | 03 | 6 | 60.0% | C | |
| HellaSwag | 93.3% | 03 | 27 | 92.3% | C | |
| AMC_2022_23 | 46.4% | 05 | 6 | 20.0% | C | |
| HiddenMath | 52.0% | 05 | 13 | 66.7% | C | |
| Vibe-Eval | 53.9% | 05 | 8 | 42.9% | C | |
| Video-MME | 78.6% | 09 | 17 | 50.0% | C | |
| MATH | 86.5% | 10 | 71 | 87.1% | C | |
| MGSM | 87.5% | 12 | 31 | 63.3% | C | |
| MathVista | 68.1% | 17 | 39 | 57.9% | C | |
| DROP | 74.9% | 20 | 30 | 34.5% | C | |
| GSM8k | 90.8% | 26 | 48 | 46.8% | C | |
| MMLU | 85.9% | 35 | 100 | 65.7% | C | |
| MMMU | 65.9% | 38 | 63 | 40.3% | C | |
| HumanEval | 84.1% | 39 | 66 | 41.5% | C | |
| MMLU-Pro | 75.8% | 70 | 129 | 46.1% | C | |
| GPQA | 59.1% | 162 | 234 | 30.9% | C |
Preference and agent-evaluation results for the default version.
| vision style control | overall | 84 | 1179.5 | 8,902 | N/A | |
| vision | overall | 93 | 1161.5 | 8,902 | N/A | |
| vision style control | overall | 110 | 1120.0 | 16,734 | N/A | |
| vision | overall | 113 | 1088.4 | 16,734 | N/A | |
| text style control | overall | 190 | 1351.2 | 55,606 | N/A | |
| text | overall | 207 | 1319.3 | 55,606 | N/A | |
| text style control | overall | 227 | 1323.9 | 79,138 | N/A | |
| text | overall | 248 | 1273.6 | 79,138 | N/A |
Provider-specific output speed and catalog latency for Gemini 1.5 Pro. Runtime does not affect the capability score.
| 85 tok/s | 0.7 s | 2.1M | 8.2K |
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
Official vendor API pricing appears first, followed by individual provider offers.
The default version has no current input or output token prices.
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
Available versions of this model. The score column identifies the version used in the overall ranking.
| Gemini 1.5 Pro | 20.5 | N/A | 2.1M | 8.2K | No | Proprietary |
Key information about Gemini 1.5 Pro and its available data.
5 Pro is a mid-size multimodal model optimized for a wide range of reasoning tasks. It can process large amounts of data at once, including 2 hours of video, 19 hours of audio, codebases with 60,000 lines of code, or 2,000 pages of text.
Data as of 2026-08-11.
Open a comparison with the three ranked models immediately above and below this model.
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
Common questions about Gemini 1.5 Pro.
Gemini 1.5 Pro's default version was released on May 1, 2024.
No official standard PAYG price is currently available for Gemini 1.5 Pro.
Gemini 1.5 Pro was created by Google.
The default version has a 2.1M token context window.
No. The default version is not marked as having publicly available weights.
No provider offering is currently linked to the default version.
Nearby ranked alternatives include Grok 2, Nova Pro, Min istral 3.