Compare models
Pick up to three models.
 Claude Haiku 4.5 |  Kimi K3 |  DeepSeek V4 Pro | |
|---|---|---|---|
| AA Intelligence Index | 24 | 57 | 44 |
| Context window | 200K tokens | 1M tokens | 1M tokens |
| Input $/MTok | $1 / MTok | $3 / MTok | $0.435 / MTok |
| Output $/MTok | $5 / MTok | $15 / MTok | $0.87 / MTok |
| Modalities | text, image, code | text, image, code | text, code |
| API | Yes | Yes | Yes |
| Open source | No | Yes | Yes |
Intelligence Index scores are sourced from Artificial Analysis, an independent evaluator — not Timace's own judgment.