Kimi K2.5 vs Qwen3 VL 8B Instruct
Moonshot AI vs Alibaba — benchmarks, pricing, and capabilities side by side.
- •Kimi K2.5 has the higher intelligence index (87.9 vs 43)
- •Qwen3 VL 8B Instruct is cheaper ($0.08 vs $0.40 per 1M input)
- •Qwen3 VL 8B Instruct is faster
- •Kimi K2.5 has a larger context window (262K)
| Kimi K2.5 | Qwen3 VL 8B Instruct | |
|---|---|---|
| Intelligence index | 87.9 | 43 |
| Developer | Moonshot AI | Alibaba |
| Type | Multimodal | Multimodal |
| Access | Open weights | Open weights |
| Context window | 262,144 tokens | 256,000 tokens |
| Input price | $0.40 / 1M | $0.08 / 1M |
| Output price | $1.90 / 1M | $0.50 / 1M |
| Speed | 35 tok/s | 145 tok/s |
| Released | January 27, 2026 | October 14, 2025 |
| Parameters | 1T (32B active) | — |
| Input modalities | Text, Image | Image, Text |
| Output modalities | Text | Text |
Shared benchmarks
Kimi K2.5
Qwen3 VL 8B Instruct
GPQA Diamond
87.9
42.7
Humanity’s Last Exam
29.4
2.9