Qwen/Qwen3-VL-235B-A22B-Thinking vs moonshotai/Kimi-K2.5
moonshotai/Kimi-K2.5 is cheaper on output tokens. Choose Qwen/Qwen3-VL-235B-A22B-Thinking or moonshotai/Kimi-K2.5 based on the trade-off between cost, context, and the benchmarks that matter for your use case.
| Spec | Qwen/Qwen3-VL-235B-A22B-Thinking | moonshotai/Kimi-K2.5 |
|---|---|---|
| Provider | SiliconFlow | SiliconFlow |
| Input / 1M tokens | $0.45 | $0.45 |
| Output / 1M tokens | $3.50 | $2.25 |
| Context window | 262K | 262K |
| Parameters | — | 1T |
| Open weights | No | Yes |
| Released | Oct 2025 | Jan 2026 |
FAQ
Pricing is indicative — confirm with the provider before production use. Highlighted values indicate the better figure for that row.