Claude Sonnet 4.6 vs Gemini 2.5 Pro
兩個模型整體表現接近
Arena ELO
Claude Sonnet 4.6
價格較低
Gemini 2.5 Pro
速度較快
Claude Sonnet 4.6
詳細比較
| 類別 | Claude Sonnet 4.6 | Gemini 2.5 Pro |
|---|---|---|
| Provider | Anthropic | |
| Context Window | 1M | 1M |
| Input Price ($/1M) | $3 | $1.25 勝出 |
| Output Price ($/1M) | $15 | $10 勝出 |
| Arena ELO | 1440 勝出 | 1360 |
| Coding | 79.6% SWE-bench | 75.6% LiveCodeBench |
| Reasoning | 72.5% OSWorld | 84.6% GPQA Diamond |
| Speed (tokens/sec) | ~80 t/s 勝出 | ~60 t/s |
| Open Source | No | No |
常見問題
Which is cheaper, Claude Sonnet 4.6 or Gemini 2.5 Pro?
Gemini 2.5 Pro is approximately 38% cheaper than Claude Sonnet 4.6 per million tokens.
Which is better for coding, Claude Sonnet 4.6 or Gemini 2.5 Pro?
Claude Sonnet 4.6 scores 79.6% SWE-bench on coding benchmarks, while Gemini 2.5 Pro scores 75.6% LiveCodeBench.
Which has a larger context window?
Claude Sonnet 4.6 supports 1M context, while Gemini 2.5 Pro supports 1M.
Is Claude Sonnet 4.6 or Gemini 2.5 Pro open source?
Claude Sonnet 4.6 is proprietary. Gemini 2.5 Pro is proprietary.
Which ranks higher on the Chatbot Arena leaderboard?
Claude Sonnet 4.6 ranks higher with ELO 1440 vs 1360.
相關文章
Anthropic extends Claude Fable access after GPT-5.6
Claude's J-space is not consciousness, but it matters
Claude Code MCP setup that connects real tools
Google’s Gemini Enterprise Agent Platform makes agents a cloud servic…
Google’s Android Bench update exposes Gemini’s gap
Claude Reflect turns usage into retention