Gemini 1.0 Pro vs Gemini 1.0 Ultra: Benchmark Comparison
Detailed comparison of Gemini 1.0 Pro and Gemini 1.0 Ultra covering benchmarks, pricing, context window, and compliance.
Key Specifications
| Specification | Gemini 1.0 Pro | Gemini 1.0 Ultra |
|---|---|---|
| Vendor | ||
| Version | 1.0-pro | 1.0-ultra |
| Release Date | 2023-12-06 | 2023-12-06 |
| Context Window | 32768 tokens | 32768 tokens |
| Input Modalities | text | text, image |
| Output Modalities | text | text |
| License | Proprietary | Proprietary |
| SOC2 | ✓ | ✓ |
| HIPAA | ✗ | ✗ |
| GDPR | ✓ | ✓ |
| ISO 27001 | ✓ | ✓ |
Benchmark Results
| Benchmark | Gemini 1.0 Pro | Gemini 1.0 Ultra | Winner |
|---|---|---|---|
| ARC | 94.1 | 95.7 | Gemini 1.0 Ultra |
| BBH | 79.1 | 83.8 | Gemini 1.0 Ultra |
| GPQA | 47.7 | 46.9 | Gemini 1.0 Pro |
| GSM8K | 82.4 | 85.9 | Gemini 1.0 Ultra |
| HUMANEVAL | 76.4 | 81.7 | Gemini 1.0 Ultra |
| IFEVAL | 75.1 | 81.6 | Gemini 1.0 Ultra |
| MATH | 58.2 | 57.1 | Gemini 1.0 Pro |
| MMLU | 84.5 | 80.3 | Gemini 1.0 Pro |
| MUSR | 57.8 | 64.3 | Gemini 1.0 Ultra |
| WINOGRANDE | 84.5 | 80.4 | Gemini 1.0 Pro |
Pricing Comparison
| Tier (per Mtok) | Gemini 1.0 Pro | Gemini 1.0 Ultra |
|---|---|---|
| Input | $0.5 | $7 |
| Output | $1.5 | $21 |
| Cache Read | $0 | $0 |
| Cache Write | $0 | $0 |
Gemini 1.0 Pro против Gemini 1.0 Ultra
Обзор модели
Gemini 1.0 Pro and Gemini 1.0 Ultra are both notable options in the AI model market. This page compares their benchmarks, pricing, and compliance.
Ключевые характеристики
| Поставщик | Дата выпуска | Окно контекста | Лицензия |
|---|---|---|---|
| Google / Google | 2023-12-06 / 2023-12-06 | 32K / 32K | Proprietary / Proprietary |
Производительность
| Бенчмарк | Gemini 1.0 Pro | Gemini 1.0 Ultra | Победитель |
|---|---|---|---|
| ARC | 94.1 | 95.7 | B |
| BBH (BIG-Bench Hard) | 79.1 | 83.8 | B |
| GPQA | 47.7 | 46.9 | A |
| GSM8K (Grade School Math 8K) | 82.4 | 85.9 | B |
| HumanEval | 76.4 | 81.7 | B |
| IFEval | 75.1 | 81.6 | B |
| MATH | 58.2 | 57.1 | A |
| MMLU (Massive Multitask Language Understanding) | 84.5 | 80.3 | A |
| MUSR | 57.8 | 64.3 | B |
| WinoGrande | 84.5 | 80.4 | A |
Сравнение цен
| Вход | Выход | Чтение кэша | Запись кэша |
|---|---|---|---|
| — / — | — / — | — / — | — / — |
за миллион токенов — A / B
Сильные стороны & Слабые стороны
Gemini 1.0 Pro
- ✅ MMLU score 84.5, strong knowledge reasoning.
- ⚠️ 闭源专有模型，不支持自托管。
Gemini 1.0 Ultra
- ✅ MMLU score 80.3, strong knowledge reasoning.
- ✅ HumanEval 81.7, excellent code generation.
- ✅ GSM8K 85.9, robust math reasoning.
- ✅ 支持文本、图像、音频多模态输入。
- ⚠️ 闭源专有模型，不支持自托管。
Мнение редактора
Gemini 1.0 Pro and Gemini 1.0 Ultra each have their strengths. Choose based on workload (code, long context, vision), referencing the tables above.
ЧЗВ
Which model is better for coding tasks?
Refer to the HumanEval benchmark table; the model with a higher score is better suited for coding tasks.
Which model is cheaper?
Refer to the pricing comparison table above; the model with lower input/output prices is more cost-effective.
Which has a longer context window?
Refer to the key specifications table; the model with a larger context window is better for long documents.
Ссылки
Editor's Take
See Editor's Take section.