Is Kimi 3 the same model as Kimi K3?

Yes — Kimi 3 and Kimi K3 are two ways of writing the same model, and this page covers both. The string the API expects is the hyphenated one, `kimi-k3`, with no date suffix and no version decimal.

What is the Kimi K3 context window?

Moonshot documents 1,048,576 tokens for Kimi K3, with output defaulting to 131,072 tokens and raisable to the full 1,048,576. The FAQ also states the rate is flat with no tiering by context length, so a long prompt costs the same per token as a short one.

Can I turn off thinking on Kimi K3?

No. Moonshot's own FAQ answers this with a flat no — Kimi K3 always thinks. If the reasoning is taking too long or costing too much, drop the reasoning effort to low rather than looking for a switch. The K2-era thinking parameter is not a Kimi K3 parameter at all and should be removed when you migrate.

What reasoning effort levels does Kimi K3 support?

Kimi K3 supports three — low, high and max — and defaults to max, the most expensive rung. There is no medium. Choose the rung before the conversation starts, because Moonshot warns that switching it mid-session invalidates prefix-cache hits.

Does Kimi K3 accept temperature and top_p?

No. Moonshot fixes temperature at 1.0, top_p at 0.95, and the count, presence-penalty and frequency-penalty parameters at their defaults; sending any other value returns an error, so leave all five out of the request. Steer Kimi K3 with prompting and the reasoning-effort rung instead.

Can I send an image URL to Kimi K3?

No — Moonshot documents that public image URLs are not supported. Send the image as a base64 data URI in an image part, or reference a file previously uploaded to Moonshot. Video is supported the same way, and the message content has to be an array of parts rather than a plain string.

How does Kimi K3 compare with Claude Opus 5?

They land at effectively the same rate on reAPI, so the choice is about shape rather than price. Kimi K3 has open weights, always reasons, hands its reasoning text back to you, and takes video. Claude Opus 5 is closed, lets thinking be switched off at lower effort, never returns raw reasoning, and takes PDF instead of video. The versus table on this page has the row-by-row detail.

Is Kimi K3 free or paid?

Paid. Usage is billed pay-as-you-go in USD from your api.reapi.ai token balance, based on input, output, and any applicable cache-token rates. See the live per-1M-token rates in the pricing table on this page; failed requests are not charged.

Does Kimi K3 require a monthly subscription?

No. There is no monthly subscription. Add funds to your api.reapi.ai token balance and pay in USD only for the input, output, and any applicable cache tokens each request uses. The live per-1M-token rates are in the pricing table on this page.

How do I get a Kimi K3 API key?

Sign up at api.reapi.ai and create a key under API Keys, then pass it as a Bearer token in the Authorization header. Chat keys and their token balance are separate from the reapi.ai image and video gateway.

Kimi K3 — 1M Context, Open-Weight Flagship

Kimi K3 is Moonshot's flagship model for long-horizon coding and end-to-end knowledge work. Kimi K3 reads a 1,048,576-token context window, always reasons, and takes a three-rung reasoning-effort dial that runs from low to max. Native vision covers images and video, the weights are open, and on reAPI Kimi K3 bills below Moonshot's published per-token rate on both input and output.

Kimi K3modelkimi-k3

Kimi K3 playground

Chat models run on the api.reapi.ai gateway, which has its own console and its own key. Open the console to run Kimi K3 through the OpenAI-compatible Chat Completions surface.

Open console Token rates

What is the Kimi K3 API?

Kimi K3 is Moonshot AI's open-weight flagship model for long-horizon coding and end-to-end knowledge work, served on reAPI through an OpenAI-compatible chat completions endpoint. It carries a 1,048,576-token context window and can return up to that entire window in a single response — the default output cap is 131,072 tokens — always reasons at low, high or max effort, and accepts native image and video input alongside tool calling and strict JSON Schema output. Input and output tokens are billed separately per million tokens, and context caching kicks in automatically on repeated prefixes.

What you can build with this model

Real-world workflows and production use cases you can build and ship with this model.

Kimi K3 fixing a layout defect from a screenshot of the broken page

Long-horizon coding, with screenshots in the loop

Moonshot positions Kimi K3 for long-horizon coding, and describes it running engineering tasks with minimal human supervision — working through large codebases and coordinating terminal tools. What separates Kimi K3 from a text-only coding model is that vision is native rather than bolted on: it takes screenshots and visual feedback as input, which Moonshot calls out specifically for game development, frontend engineering and CAD. That closes the loop a text-only model cannot close, because the model can look at what it just built and judge whether the layout is actually wrong before it edits again.

Read the API docs

Kimi K3 reading a large document set and producing a structured deliverable

A million tokens of knowledge work with no chunking layer

The Kimi K3 context window is 1,048,576 tokens, and Moonshot's own FAQ says the rate is flat with no tiering by context length — so filling the window costs what any other token costs. A contract set, a research bundle or a year of filings goes to Kimi K3 whole, which removes the retrieval layer whose only job was working around a small context. Context caching then does the repeat work: it is automatic on Kimi K3, with no cache id, no TTL and no extra parameter, so a stable prefix in front of a changing question is the pattern to build around.

Kimi K3 running an agentic loop with tools loaded partway through

Agents that are made to call tools, not talked into it

Two tool-calling controls arrived with this generation. A required tool choice forces at least one call on the turn, which is how you stop an agent from answering from memory when it was supposed to look something up — the K2 models reject that setting, Kimi K3 accepts it. And a tool can be introduced partway through a conversation by putting its full definition in a system message, so a large tool inventory does not have to be declared up front on every request. Structured outputs and a JSON Schema keep whatever Kimi K3 returns parseable at the end of the loop.

Pricing

Credit-based — 1 credit = $0.001 USD. Pay only for completed generations.

First test cost

$2.52,500 credits

One sample at the cheapest tier (1M tokens).

Testing budget guide

Add credits

$10

≈ 4 tests

$50

≈ 20 tests

$100

≈ 40 tests

Model	Category	Price
Token pricing	Input	$2.5 1M tokens
Token pricing	Output	$12 1M tokens

Why reAPI

Below the published rate, both directions

Moonshot publishes a per-million-token rate for Kimi K3 input and output. On reAPI, Kimi K3 bills below it on both, and a full fifth below on output — no subscription underneath it, no minimum commitment, pay-as-you-go in USD against your gateway balance. The exact numbers are in the pricing table on this page.

OpenAI-compatible drop-in

Kimi K3 is served from a standard `/v1/chat/completions` endpoint, and Moonshot's own quickstart drives the model through the OpenAI SDK — so the compatibility is the vendor's design, not a shim. Adopting Kimi K3 is a base URL, a key and a model string. No SDK swap, and nothing about Kimi K3 needs a bespoke client.

One key across Kimi, Claude, GPT and Gemini

The same api.reapi.ai key that calls Kimi K3 calls the Claude, GPT and Gemini families too. Putting Kimi K3 head to head with a closed frontier model, or failing over when one provider has a bad hour, is a model string rather than a second integration, a second invoice and a second set of credentials.

Kimi K3 vs Claude Opus 5

On reAPI these two land at effectively the same per-token rate, which makes price the least interesting part of the comparison. What differs is shape: Kimi K3 is the open-weight model that always reasons, hands its reasoning text back and takes video input, while Claude Opus 5 is the closed model with a finer effort ladder that keeps its reasoning to itself. Two of these rows are the ones that break code carried over from another provider — thinking cannot be switched off on Kimi K3, and its reasoning text has to be echoed back in multi-turn histories.

Capability

Kimi K3 on reAPI

Claude Opus 5

Weights

Open — Moonshot calls it the first open-source model in the three-trillion-parameter class

Closed — API access only

Context and output limits

1,048,576-token context; output defaults to 131,072 and can be raised to the whole window

1M-token context; 128k maximum output, fixed

Thinking

Always on and cannot be disabled — lower the effort instead

On by default, and can be disabled at effort high or lower

Effort ladder

Three rungs — low, high, max — defaulting to max

Five rungs from low through max, defaulting to high

Reasoning text

Returned alongside the answer, and required back in the next request

Raw reasoning never returned; a readable summary on request

Visual input

Images and video, as base64 or an uploaded file — public URLs rejected

Images and PDF; no video input

Comparison reflects behavior documented by each vendor at the time of writing. Rates are described as relationships rather than amounts; the live numbers for Kimi K3 are in the pricing table above.

Ship Kimi K3 in three steps

step 01
Create an account and key on api.reapi.ai
Sign up at api.reapi.ai, open the console and generate an API key. That one key reaches Kimi K3 and every other model on the gateway, billed from a single balance.
Open
step 02
Point your client at the gateway
Set your base URL to api.reapi.ai and your key to the one you just created. Any client that already speaks OpenAI Chat Completions works unchanged — no SDK swap is needed to reach Kimi K3.
Open
step 03
Send `kimi-k3` as the model
Set the model field to `kimi-k3` and post your messages. Turn on streaming for long outputs, drop any temperature or top_p you were sending, and pick a reasoning effort — Kimi K3 always thinks, and the default rung is the most expensive one.
Open

docs/api/kimi-k3

API reference

Drop-in code and the full parameter table.

View full reference

curl https://api.reapi.ai/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "kimi-k3",
    "group": "default",
    "messages": [
      { "role": "user", "content": "Hello" }
    ],
    "reasoning_effort": "high",
    "stream": true
  }'

Parameter validation failed

Rejected synchronously with the failing field named. Check enum values, ranges, and URL fields — nothing is charged.

Authentication or balance issue

401 means a missing or invalid Bearer key; 402 means the reserve exceeds your balance. Manage keys and credits in the console.

Content or material rejected

Reference material and outputs pass automated moderation. Rejected tasks fail with a clear error and a full refund.

Task failed or timed out

A task that reaches a failed state is never charged — the reserve refunds automatically. Keep the task id and retry when ready.

Frequently asked questions

Common questions about this model.

Kimi K3 is billed pay-as-you-go in USD per token against your api.reapi.ai balance, with separate input and output rates and no subscription. Both sit below Moonshot's published per-token rate, output by a full fifth. Current numbers are in the pricing table on this page.