Kimi models in Taskade are the reliable, efficient end of the catalog. They build steadily and hold long context well.
Kimi models available today
| Model | Good at | Speed | Intelligence | Cost | Availability |
|---|---|---|---|---|---|
| Kimi K3 | Long-context agentic reasoning | Slow | Frontier | High | Max and Enterprise |
| Kimi K2.7 Code | Agentic coding | Medium | High | Medium | Every paid plan |
| Kimi K2.6 | Open-weight generalist | Medium | High | High | Every paid plan |
How to pick within Kimi
- K3 when the request carries a lot of context at once, such as a long document or a large existing app.
- K2.7 Code for coding-shaped work specifically.
- K2.6 as the general-purpose open-weight option.
This family is a good default when you want predictable behavior over peak scores.
Set it on a step
Open the model menu on an agent, a chat, or an automation step, then choose from Custom Engine in the builder or All Models elsewhere. The choice applies only to that surface.
If you do not want to pin a model, a rung sets a speed-to-depth trade-off and follows the catalog forward as models change.
How Kimi performs
We build the same apps on every family and publish what happened. See the Kimi build results for the round-by-round record.
Next steps
- Compare every family in the full catalog.
- Read how Taskade picks a model for Auto, rungs, and Infinity Mode.