Kimi K3.1 Pro API pricing on IteraCompute
Every IteraCompute Kimi K3.1 Pro rate we track, compared against 20 providers serving the same model.
The cheapest Kimi K3.1 Pro input price we currently track is $0.240/1M on OpenRelay.
The cheapest Kimi K3.1 Pro output price is $2.40/1M on Liquid Inference.
IteraCompute Kimi K3.1 Pro rates
Prices per 1M tokens. Batch and cached-input rates are shown where the provider publishes them.
| Mode | Input | Output |
|---|---|---|
| Standard | $1.60/1M | $8.00/1M |
About Kimi K3.1 Pro
- Creator
- Moonshot AI
- Modalities
- text
Kimi K3.1 Pro on other providers
- OpenRelay$0.240/1M
- OpenRouter$0.253/1M
- Liquid Inference$1.60/1M
- AkashML$1.75/1M
- Deep Infra$1.90/1M
- Baseten$2.00/1M
- DigitalOcean$2.00/1M
- Empirio Labs$2.00/1M
- Fireworks AI$2.00/1M
- GMI Cloud$2.00/1M
- Lyceum$2.00/1M
- Morse$2.00/1M
- Novita AI$2.00/1M
- PopNow$2.00/1M
- Prime Intellect$2.00/1M
- Runpod$2.00/1M
- Together AI$2.00/1M
- HiAPI$2.15/1M
- RouterPlus$2.50/1M
About IteraCompute
IteraCompute is an inference API that serves models from Qwen, DeepSeek, GLM (Z.ai), Kimi (Moonshot AI) and MiniMax through one OpenAI-compatible endpoint, with a free tier on selected models and reserved, dedicated and custom deployments for production workloads.
Frequently Asked Questions
How much does Kimi K3.1 Pro cost on IteraCompute?
IteraCompute serves Kimi K3.1 Pro from $1.60 per 1M input tokens, across 1 tracked pricing mode. Prices are collected daily — see the table above for current input, output, and cached-input rates.
Is IteraCompute the cheapest way to run Kimi K3.1 Pro?
IteraCompute ranks #4 of 20 providers we track serving Kimi K3.1 Pro on input price. The cheapest input price right now is on OpenRelay. The lowest output price is $2.40/1M on Liquid Inference. Rates change often, and throughput, latency and context limits differ between providers — check the comparison above before committing.
Does IteraCompute offer batch or cached pricing for Kimi K3.1 Pro?
We track 1 Kimi K3.1 Pro offering from IteraCompute. Batch and cached-input rates appear in the table above when the provider publishes them; blank cells mean we have no current figure for that mode.