Kimi K3.1 Pro for code, context and careful reasoning
Give Kimi K3.1 Pro the source material, constraints and a result you can check. Work through code changes, compare documents or continue a tool conversation. $1.60 / 1M input and $8.00 / 1M output.
- 1M-token context
- Always-on reasoning
- Structured output
- Image and video input
- Streaming
Try Kimi K3.1 Pro
Start with a question you can verify. Choose a request format, inspect the response and continue with the complete conversation history.
kimi-k3.1-proPer 1M tokens: $1.60 input · $8.00 output.Blank fields are omitted. Use JSON for nested values.
Estimated from returned usage and live rates. A dash means cache lifetime or usage was not reported.
Kimi K3.1 Pro pricing
Kimi K3.1 Pro charges by input, output and cache usage. These rates come from the same current price settings used for billing.
Input, cache reads and cache writes are separate categories. Reasoning counts as output. Example sizes illustrate the calculation; actual usage determines the charge.
- Input
- $1.60Price · 1M tokens
- Output
- $8.00Price · 1M tokens
- Cached input
- $0.16Price · 1M tokens
- Cache write, 5 min
- $1.60Price · 1M tokens
- Cache write, 1 hour
- $3.20Price · 1M tokens
What is Kimi K3.1 Pro?
Kimi K3.1 Pro is a Moonshot AI reasoning model for coding, tool use and work that depends on a substantial body of source material. Its 1,048,576-token context can hold a long conversation or a carefully selected collection of files. A larger window gives you room to supply evidence; it does not guarantee that every conclusion is correct.
Kimi K3.1 Pro thinks on every request. Select low, high or max effort according to the problem, then check whether the answer meets your acceptance criteria. The default is max. Reasoning and visible answer text share the output budget, so a very small limit can finish before a useful answer appears.
Model ID: kimi-k3.1-pro · Output: text and tool calls · Native formats: Chat Completions, Responses and Messages.
Use Kimi K3.1 Pro from your application or coding assistant
Keep a familiar request format while testing your own workload.
Connect Kimi K3.1 Pro
Choose the native format your application already uses. Send kimi-k3.1-pro as the model and preserve the returned reasoning and tool history when you continue.
- 1Create an API key.
- 2Copy a request from the docs.
- 3Check the result with your own data.
Build with Kimi K3.1 Pro
Give your coding assistant the documented endpoint, your acceptance criteria and a small first task. Review the proposed request and spending limit before a paid run.
- 1Copy the setup instructions.
- 2Keep your key in an environment variable.
- 3Review before running.
What Kimi K3.1 Pro can help you do
The most useful prompts name a task, include its evidence and define a checkable result.
Explain a difficult code path
Ask Kimi K3.1 Pro to trace a request through several files, identify the assumption that fails and propose a focused change. Include the failing input so the explanation has something concrete to resolve.
Compare evidence across documents
Give Kimi K3.1 Pro the relevant policies, notes or reports. Request the supporting passage for each conclusion, and ask it to separate contradictions from information that is simply missing.
Read screenshots and clips
Combine visual material with a specific question: which chart changed, which control is missing, or what happened between two frames? Media support follows the selected request format; see the docs before preparing large inputs.
Continue a tool conversation
Kimi K3.1 Pro can request a function call and reason over its result. Your application validates arguments and executes the function. Keep that control in your application, especially when a tool changes external data.
Practical uses for Kimi K3.1 Pro
Review a risky retry
Show Kimi K3.1 Pro the worker, the external operation and the crash point. Ask for a failure analysis, a minimal correction and regression cases that exercise repeated delivery.
Extract records into a schema
Use Kimi K3.1 Pro with a JSON Schema for the fields you need. Specify how to represent missing values, and validate the returned object before saving it to your application.
Plan a bounded migration
Provide the current interfaces, dependencies and rollout constraints. Ask Kimi K3.1 Pro for small, verifiable steps with a rollback condition, rather than a broad rewrite that is difficult to assess.
Start with Kimi K3.1 Pro
- 01
Add credit
Create an account and add the amount you need. Your credit does not expire.
- 02
Choose a format
Choose Chat Completions, Responses or Messages. Use the exact model ID kimi-k3.1-pro and copy a matching example from the API docs.
- 03
Try a small task
Give Kimi K3.1 Pro a question with an answer you can check. Set an output budget that allows for reasoning.
- 04
Inspect and continue
Review the Kimi K3.1 Pro output and usage. Preserve the complete assistant and tool history when continuing the conversation in your own application.
Kimi K3.1 Pro capabilities at a glance
Match the format and controls to the task before you integrate.
- Context window
- 1,048,576 tokens
- Reasoning effort
- low, high or max; always enabled
- Structured output
- JSON Schema in all three formats
- Media inputs
- Format-specific image and video blocks
- Conversation history
- Replay complete reasoning and tool items
These are documented model capabilities. Evaluate accuracy and response time on your own workload.
Plan a successful Kimi K3.1 Pro request
Allow room for reasoning
Kimi K3.1 Pro includes thinking in its output limit. Use a reasonable budget for the task and inspect the completion status. Reaching the token limit is different from receiving a complete answer.
Prepare the correct media format
Kimi K3.1 Pro does not fetch arbitrary public image URLs. Use supported inline media or an accepted file reference. Chat accepts images and video; Responses and Messages have their own content schemas.
Keep a stable conversation prefix
Reuse the same instructions and cache key across a session. Existing cached context keeps its original lifetime when refreshed. A requested lifetime does not prove what was written; inspect the returned usage.
Verify important conclusions
Ask Kimi K3.1 Pro for source passages or executable checks. Review generated code, extracted amounts and factual claims before using them to make an external change.
Kimi K3.1 Pro on SeedRouter
See the rates here
Compare Kimi K3.1 Pro input, output and cache rates before you send a request. The pricing table updates from current billing settings.
Pay for usage
Use Kimi K3.1 Pro from a prepaid balance. There is no monthly subscription, and failed requests are not charged.
Preserve the complete history
Keep Kimi K3.1 Pro assistant messages, reasoning fields and tool identifiers unchanged between turns. Rebuilding history from visible answer text alone can remove information needed for the next step.
Prefer a concrete output contract
For extraction and classification, define the object shape and required fields. Give Kimi K3.1 Pro instructions for ambiguous cases, then validate the result in your application.
Compare Kimi K3.1 Pro with other models
See live prices and task-specific controls before choosing.
Call DeepSeek V4.1 Flash with the OpenAI or Anthropic SDK. 1M-token context, thinking on or off, image input, live peak and off-peak pricing and a playground.
View modelCall Kimi K3 with the OpenAI or Anthropic SDK. 1M-token context, always-on reasoning with low, high or max effort, live per-token pricing and a playground.
View modelCall Claude Fable 5 with the Anthropic or OpenAI SDK. Deep reasoning, 1M-token context, forced tool use, live per-token pricing and a playground to test it.
View modelCall Claude Fable 5.1 with the Anthropic or OpenAI SDK. Deep reasoning, 1M-token context, cheap cached reads, live pricing and a playground to try it today.
View modelTry Claude Haiku 5.5 for classification, extraction and summaries. See live token prices, test your prompts, and connect your app with the Messages API.
View modelCall Claude Opus 5.5 with the Anthropic or OpenAI SDK. 1M-token context, adaptive thinking, live per-token pricing and a playground you can try right now.
View modelKimi K3.1 Pro guides
Understand token billing, cache behavior and practical integration choices before scaling your workload.
Kimi K3.1 Pro questions
What is Kimi K3.1 Pro?+
Kimi K3.1 Pro is a Moonshot AI reasoning model for coding, tool workflows and work with long source material. It accepts text and format-specific visual inputs, then returns text or tool calls.
Is Kimi K3.1 Pro free?+
Kimi K3.1 Pro requests on this service use prepaid credit at the live rates above. There is no subscription requirement. Failed requests are not charged.
Can Kimi K3.1 Pro turn thinking off?+
No. Kimi K3.1 Pro always reasons. Choose low, high or max effort; max is the default. Reasoning tokens count as output even when the final answer is short.
How much context can Kimi K3.1 Pro use?+
Kimi K3.1 Pro has a 1,048,576-token context window. Input must fit inside that window. The service can reduce the requested generation budget to the remaining space; include only the material your task needs.
Which request formats does Kimi K3.1 Pro support?+
Kimi K3.1 Pro has native Chat Completions, Responses and Messages contracts. Their field names and media types differ. Use the matching docs and keep complete native history when continuing.
How much does Kimi K3.1 Pro cost?+
Kimi K3.1 Pro currently starts at $1.60 / 1M input and $8.00 / 1M output. The live table also separates cached reads and cache writes by lifetime. Your usage record shows the charge for each request.
Try Kimi K3.1 Pro on your own task
Give Kimi K3.1 Pro a question you can verify. Review the answer and usage.