Coding
Aider-polyglot
Benchmark de edicion de codigo en multiples lenguajes.
6 modelos publicaron score
| # | Modelo | Empresa | Score |
|---|---|---|---|
| 1 | Claude Opus 4.5 | Anthropic | 89.4 |
| 2 | GPT-5.2 | OpenAI | 88.0 |
| 3 | Grok 4 | xAI | 79.6 |
| 4 | DeepSeek V3.2 | DeepSeek | 74.5 |
| 5 | Doubao Seed 2.0 Pro | ByteDance | 54.2 |
| 6 | GLM-4.6 | Zhipu AI | 39.1 |