Head to head
Qwen 3.8-Max vs Kimi K3
Which is better, Qwen 3.8-Max or Kimi K3?
Qwen 3.8-Max finishes ahead of Kimi K3 on the index, 85.1 to 85.1. It is the stronger of the two on Speed, while Kimi K3 still leads on Control. Both are scored on the same rubric, in the same week, and neither placing is sponsored.
Updated
Qwen 3.8-Max
18 of 73
85.1
Alibaba's flagship, and since August its weights are published too.
Kimi K3
20 of 73
85.1
One of the strongest open-weights models, and it reads a million tokens without blinking.
The four scores, side by side
| Score | Qwen 3.8-Max | Kimi K3 | Difference |
|---|---|---|---|
| Craft | 88 | 88 | — |
| Speed | 78 | 74 | +4 |
| Control | 86 | 91 | +5 |
| Value | 80 | 78 | +2 |
| Index score | 85.1 | 85.1 | +0.0 |
Which to pick
Pick Qwen 3.8-Max unless Control is what decides it for you, which is exactly where Kimi K3 is the better answer.
Qwen 3.8-Max
3.8-Max is the top of Alibaba's line, and in August the company reversed course and published its weights as a 2.4-trillion-parameter sparse model. It sits in the top group on both independent leaderboards, covers more languages than almost anything else here, and a stable API snapshot from September ended the preview churn. The licence is permissive for most users but requires a separate agreement from large companies selling it as a service.
Where it shines
- Broadest language coverage of any frontier model
- Weights published as a 2.4-trillion-parameter sparse model
- API priced well below Western flagships
Kimi K3
Moonshot published K3 at a very large scale, with native vision, an explicit thinking mode and a million-token window, and in October it still ranks among the top open models on both independent leaderboards. Two caveats matter: the licence restricts commercial use above a revenue threshold, so it is open weights rather than open source, and the hosted API costs three dollars in and fifteen out, priced like a Western frontier model rather than a Chinese one.
Where it shines
- Million-token context handled properly rather than nominally
- Native vision and an explicit thinking mode
- Among the highest human-preference ratings of any open model