Skip to content
ShilpBot
AI Models

MoonshotAI Kimi K3.1 Pro: Pricing, Specs and Limits

MoonshotAI Kimi K3.1 Pro: flagship open-weight multimodal reasoning model, 1M context, 0.24 in / 8.40 out per 1M tokens. Specs, limits and test prompts.

By ShilpBot Team · Published · 3 min read

Kimi K3.1 Pro is Moonshot AI's flagship open-weight multimodal reasoning model and a post-trained successor to Kimi K3, with a 1M-token context window. The listing says it is built for long-horizon coding and 3D work, among other uses.

Its pricing is unusual: only $0.24 per million input tokens but $8.40 per million output tokens. Tasks that read a lot and write little are cheap, while tasks that produce long outputs are expensive.

Prices and specifications checked against the public OpenRouter catalogue on 2026-11-16. See the live model page for current limits.

MoonshotAI Kimi K3.1 Pro quick facts

DetailValue
DeveloperMoonshot AI
Model IDmoonshotai/kimi-k3.1-pro
Added to the catalogue2026-10-23
TypeFlagship open-weight multimodal reasoning model
ParametersNot listed
Context windowabout 1M tokens (1,048,576)
Maximum output943,718 tokens
Input typestext, image, video
Price0.24 in / 8.40 out per 1M tokens (cached input 0.20)

Feature support from the catalogue: Tool calling: listed; JSON / structured output: listed; Adjustable reasoning effort: listed.

What MoonshotAI Kimi K3.1 Pro is good for

  • Read-heavy tasks such as long-document questions and code review with short answers.
  • Long-horizon coding agents that benefit from open-weight models.
  • Image and video understanding alongside text.

Limits of MoonshotAI Kimi K3.1 Pro

  • Output tokens cost 35 times as much as input tokens, so reasoning-heavy answers can be pricey.
  • Cached input ($0.20) is almost the same as uncached ($0.24), so caching saves little here.
  • Cap output length on agent loops.

How MoonshotAI Kimi K3.1 Pro compares

  • Reflection Beam 1.0: Both have a 1,048,576-token context window; both cap output at 943,718 tokens; it also accepts image and video input; it costs 0.24 input and 8.40 output per million tokens against 0.30 and 1.00.
  • DeepSeek V4.1 Flash 1017: Both have a 1,048,576-token context window; its output cap is 943,718 tokens against 393,216; it also accepts video input; it costs 0.24 input and 8.40 output per million tokens against 0.15 and 0.60.
  • MiniMax M3.1 Flash: Both have a 1,048,576-token context window; its output cap is 943,718 tokens against 524,288; it costs 0.24 input and 8.40 output per million tokens against 0.15 and 0.60.

Calling MoonshotAI Kimi K3.1 Pro from code

Use the model ID moonshotai/kimi-k3.1-pro with any OpenAI-compatible client pointed at OpenRouter:

curl https://openrouter.ai/api/v1/chat/completions \
  -H "Authorization: Bearer $OPENROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "moonshotai/kimi-k3.1-pro", "messages": [{"role": "user", "content": "Explain what you can do in two sentences."}]}'

Prompts that show what MoonshotAI Kimi K3.1 Pro can do

  1. Ask a precise question about a 200K-token document and compare cost with the same question on a model with cheaper output.
  2. Give it a coding task and ask for a short plan before any code.
  3. Measure total cost of a 20-step agent run.

What MoonshotAI Kimi K3.1 Pro costs

The catalogue lists 0.24 per million input tokens and 8.40 per million output tokens, with cached input at 0.20 per million. Sending 1 million input tokens and receiving 100,000 output tokens would cost about 1.08. A single coding-agent step with a 150,000-token prompt and a 10,000-token answer would cost about $0.12 before caching. If your client reuses the same prompt prefix, cached input at 0.20 per million tokens can cut the input side of that bill by about 17 percent. Prices are set by the provider, can change, and may differ between hosting providers, so check the live model page before you budget.

Keep reading

Compare every paid model by price in our AI model pricing roundup, or see the free models roundup. New to terms like mixture-of-experts? Start with how neural networks learn.

Frequently asked questions

Why is Kimi K3.1 Pro's output so much costlier than input?

The catalogue lists $0.24 per million input tokens and $8.402 per million output tokens. Providers sometimes price reasoning output high relative to input. Estimate your own input-to-output ratio before choosing.

What are the cost and limits of MoonshotAI Kimi K3.1 Pro?

As of 2026-11-16 the catalogue lists 0.24 per million input tokens and 8.40 per million output tokens. About 1M tokens (1,048,576) context window, up to 943,718 output tokens, model ID moonshotai/kimi-k3.1-pro. Prices can change, so check the live model page.

#MoonshotAI Kimi K3.1 Pro#Moonshot AI model#AI model pricing#AI API pricing#AI chat model#reasoning model#OpenRouter