Gemini 1.5 Pro vs Claude 3 Opus: Benchmark Comparison
Detailed comparison of Gemini 1.5 Pro and Claude 3 Opus covering benchmarks, pricing, context window, and compliance.
Key Specifications
| Specification | Gemini 1.5 Pro | Claude 3 Opus |
|---|---|---|
| Vendor | anthropic | |
| Version | 1.5-pro | 3-opus |
| Release Date | 2024-02-15 | 2024-03-04 |
| Context Window | 2e+06 tokens | 200000 tokens |
| Input Modalities | text, image, audio, video | text, image |
| Output Modalities | text | text |
| License | Proprietary | Proprietary |
| SOC2 | ✓ | ✓ |
| HIPAA | ✗ | ✗ |
| GDPR | ✓ | ✓ |
| ISO 27001 | ✓ | ✗ |
Benchmark Results
| Benchmark | Gemini 1.5 Pro | Claude 3 Opus | Winner |
|---|---|---|---|
| ARC | — | 94.5 | Claude 3 Opus |
| BBH | 84 | 81.6 | Gemini 1.5 Pro |
| GPQA | — | 46 | Claude 3 Opus |
| GSM8K | 91.7 | 91.8 | Claude 3 Opus |
| HUMANEVAL | 71.9 | 83.3 | Claude 3 Opus |
| IFEVAL | — | 79.2 | Claude 3 Opus |
| MATH | 58.5 | 42.7 | Gemini 1.5 Pro |
| MMLU | 85.9 | 81.2 | Gemini 1.5 Pro |
| MUSR | — | 53.3 | Claude 3 Opus |
| WINOGRANDE | — | 81.9 | Claude 3 Opus |
Pricing Comparison
| Tier (per Mtok) | Gemini 1.5 Pro | Claude 3 Opus |
|---|---|---|
| Input | $1.25 | $15 |
| Output | $5 | $75 |
| Cache Read | $0.3125 | $0 |
| Cache Write | $1.25 | $0 |
Gemini 1.5 Pro 対 Claude 3 Opus
モデル概要
Gemini 1.5 Pro and Claude 3 Opus are both notable options in the AI model market. This page compares their benchmarks, pricing, and compliance.
主要仕様
| ベンダー | リリース日 | コンテキストウィンドウ | ライセンス |
|---|---|---|---|
| Google / Anthropic | 2024-02-15 / 2024-03-04 | 2000K / 200K | Proprietary / Proprietary |
ベンチマークパフォーマンス
| ベンチマーク | Gemini 1.5 Pro | Claude 3 Opus | 勝者 |
|---|---|---|---|
| ARC | — | 94.5 | B |
| BBH (BIG-Bench Hard) | 84.0 | 81.6 | A |
| GPQA | — | 46.0 | B |
| GSM8K (Grade School Math 8K) | 91.7 | 91.8 | Tie |
| HumanEval | 71.9 | 83.3 | B |
| IFEval | — | 79.2 | B |
| MATH | 58.5 | 42.7 | A |
| MMLU (Massive Multitask Language Understanding) | 85.9 | 81.2 | A |
| MUSR | — | 53.3 | B |
| WinoGrande | — | 81.9 | B |
料金比較
| 入力 | 出力 | キャッシュ読み取り | キャッシュ書き込み |
|---|---|---|---|
| — / — | — / — | — / — | — / — |
100万トークンあたり — A / B
強み & 弱み
Gemini 1.5 Pro
- ✅ MMLU score 85.9, strong knowledge reasoning.
- ✅ GSM8K 91.7, robust math reasoning.
- ✅ 支持文本、图像、音频多模态输入。
- ✅ 上下文窗口 2000K，支持长文本。
- ⚠️ 闭源专有模型，不支持自托管。
Claude 3 Opus
- ✅ MMLU score 81.2, strong knowledge reasoning.
- ✅ HumanEval 83.3, excellent code generation.
- ✅ GSM8K 91.8, robust math reasoning.
- ✅ 支持文本、图像、音频多模态输入。
- ✅ 上下文窗口 200K，支持长文本。
- ⚠️ 闭源专有模型，不支持自托管。
編集者コメント
Gemini 1.5 Pro and Claude 3 Opus each have their strengths. Choose based on workload (code, long context, vision), referencing the tables above.
FAQ
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.
参考文献
Editor's Take
See Editor's Take section.