Kimi K3.1 Pro Highspeed · Gateway

What you can use today

Token prices measure the small pieces of text the model reads or produces. Capacity is the amount the service accepts in a request; check the limits below for any fallback to another model. A dash means the rate is unverified.

Exact AI Gateway model ID: moonshotai/kimi-k3.1-pro-highspeed. Gateway documents low, high and max reasoning effort, but does not state an omitted default. Native Kimi defaults to max and keeps thinking on. Check the selected service’s settings.

Available thinking settings: low, high, max. We do not yet have comparable independent results for these settings. The default applied when a request omits effort is unverified.

This is a serving option for K3.1 Pro, not a separately benchmarked base model. The publisher’s speed claim is not a measured advantage for your workload.

Gateway declares 1,048,576 context and maximum output tokens. Prompt, thinking and answer must still fit the shared context.

Per million tokens: $4 input, $0.20 cache reads, $20 output and $4 cache writes. Standard K3.1 Pro on Gateway lists $2/$0.20/$10/$2. Highspeed doubles the input and output quote, not the cache-read quote. Allow for retries and any tool fees. Models that think before answering may also charge for that work.

Provider release and specifications ↗

Check availability and plans ↗

Provider documentation ↗

Test it before switching

Give each model the same examples, tools and definition of success. Record correct outcomes, response time, retries and total cost. A cheaper token price does not guarantee a cheaper completed task.