How to Use Kimi K3: Access, API Setup, and Pricing (2026)

How to Use Kimi K3: Access, API Setup, and Pricing (2026)

Kimi K3 is Moonshot AI's new flagship model: 2.8 trillion parameters, a 1 million token context window, and native support for text, image, and video input. It launched the week of July 14, 2026, and sits above the rest of Moonshot's K-series lineup. This guide skips the launch-day hype and covers the part most coverage doesn't: how to actually get access, what it costs, and where it fits your work. For the full spec rundown, see our What Is Kimi K3 explainer.

There are two ways to use it: the consumer chat app at kimi.com, or the developer API. This guide walks through both, with a working code snippet for the API path and the real, now-published pricing numbers.

Quick Answer

• Chat app: go to kimi.com and start a conversation. K3 is the flagship model behind Kimi's consumer product.
• API: sign up at platform.kimi.ai, generate a key, and call https://api.moonshot.ai/v1 with an OpenAI-compatible request. Model ID: kimi-k3.
• Pricing (verified July 21, 2026, Moonshot's official rate card): $3.00 per million input tokens on a cache miss, $0.30 per million on a cache hit, $15.00 per million output tokens.
• Not open-weight. There's no K3 model on Moonshot's Hugging Face page as of this writing. If you need downloadable weights, K2.5 is the newest option Moonshot has actually open-sourced.

How to Access Kimi K3 (Two Paths)

1. The Consumer App: kimi.com

The fastest way to try K3 with no setup. Kimi's website runs K3 as its flagship model, so opening kimi.com and starting a chat puts you on the current model without picking anything from a dropdown. This is the right starting point if you just want to test what K3 can do before wiring up an integration.

2. The Developer API: platform.kimi.ai

For anything programmatic, coding agents, internal tools, batch jobs, you'll want the API. Here's the setup path:

1. Create an account at platform.kimi.ai.
2. Open the API Keys section of the console and generate a new key.
3. Set the key as an environment variable (MOONSHOT_API_KEY is the name Moonshot's own quickstart docs use).
4. Point your client at the base URL: https://api.moonshot.ai/v1.

One thing worth flagging if you've read older coverage of Kimi's API: the base URL has moved. Earlier documentation (and some still-circulating guides) point to api.moonshot.cn/v1. As of this writing, Moonshot's own quickstart docs at platform.kimi.ai list the base URL as api.moonshot.ai/v1. Use the .ai domain and confirm the current value in Moonshot's docs before you ship, since a launch-week platform like this can keep shifting.

The API is OpenAI-format compatible, so if you already have code calling an OpenAI-style chat completions endpoint, switching over is mostly a matter of changing the base URL, key, and model string.

Python:

import os
from openai import OpenAI

client = OpenAI(
    api_key=os.environ["MOONSHOT_API_KEY"],
    base_url="https://api.moonshot.ai/v1",
)

completion = client.chat.completions.create(
    model="kimi-k3",
    messages=[
        {"role": "user", "content": "Summarize this document and flag any open questions."},
    ],
)

print(completion.choices[0].message.content)

curl:

curl https://api.moonshot.ai/v1/chat/completions \
    -H "Content-Type: application/json" \
    -H "Authorization: Bearer $MOONSHOT_API_KEY" \
    -d '{
        "model": "kimi-k3",
        "messages": [{"role": "user", "content": "What is 1+1?"}]
    }'

Both examples come straight from Moonshot's own quickstart documentation, adjusted only to show the key as an environment variable rather than a hardcoded string.

Kimi K3 Pricing: What It Actually Costs

Earlier reporting on K3 (including our own initial coverage right after launch) had to note that Moonshot listed separate input and output pricing without publishing the exact per-token numbers. That's changed. As of this writing, Moonshot's pricing page for K3 lists:

Source: platform.kimi.ai/docs/pricing/chat-k3, verified July 21, 2026.

The cache-hit discount is the detail worth planning around. If your workload repeats large chunks of context, a long system prompt, a big document, a persistent agent scratchpad, the cache-hit rate is a tenth of the cache-miss rate. Structure your prompts so the stable, reusable parts come first and the part that changes per request comes last, which is standard practice for prompt caching in general and applies here too.

Separately, Moonshot's pricing docs note that file-related operations (upload and content extraction) are temporarily free. That's a time-limited promotion, not a permanent feature, so don't build a cost model that assumes it stays that way.

Pricing on a platform this new can change without much notice. Check Moonshot's live pricing page before you commit budget to it.

What Kimi K3 Is Good For (and What It Isn't)

Moonshot's own documentation positions K3 as the "flagship model for long-horizon coding and end-to-end knowledge work," and describes it as "well suited for programming agent scenarios." Combined with the 1 million token context window and native multimodality, text, image, and video input without a separate vision adapter, that points to a few concrete use cases:

• Long-running coding agents. The context window is large enough to hold a substantial codebase or a long multi-step task history without truncation.
• Document-heavy knowledge work. Feed it long reports, transcripts, or research bundles and ask it to reason across all of it at once.
• Multimodal input tasks. Since it handles image and video natively, you can pass visual context alongside text without a separate pipeline.

What it isn't, at least not yet:

• Not open-weight. There's no K3 repository on Moonshot's Hugging Face page. If self-hosting or fine-tuning your own copy matters, K2.5 remains Moonshot's newest actual open release.
• Not independently benchmarked. No official K3 benchmark suite has been published as of this writing, and no independent evaluator has released a verified score either. Treat any specific benchmark number attributed to K3 right now, on any site, as unconfirmed.
• Always in reasoning mode. Moonshot's docs describe a reasoning_effort request field (low, high, or max, default max) rather than a true on/off switch for reasoning. If you need the fastest, cheapest response for a simple task, set reasoning_effort to low rather than expecting a separate non-reasoning model.

Tips for Getting the Most Out of Kimi K3

• Tune reasoning_effort to the task. Moonshot's docs expose this as a top-level request field with three settings. Save max for genuinely hard, multi-step problems and drop to low for straightforward lookups or formatting tasks, since reasoning effort affects both latency and output token count (which is what you're billed for).
• Structure prompts for cache hits. Put stable content, system instructions, reference documents, tool definitions, ahead of the part of the prompt that changes each call. That's what makes the $0.30 cache-hit rate apply instead of the $3.00 cache-miss rate.
• Don't hardcode the base URL from an old tutorial. The API has already moved from api.moonshot.cn/v1 to api.moonshot.ai/v1 once. Pull the current value from Moonshot's docs rather than copying it from a guide, including this one, without checking the date.
• Keep the model string configurable. Moonshot hasn't published a single canonical model identifier across all its docs pages; confirm kimi-k3 still matches what's live in your console before shipping to production.

FAQ

How do I use Kimi K3? Through the consumer app at kimi.com for chat, or through Moonshot's API at api.moonshot.ai/v1 for programmatic access. The API uses an OpenAI-compatible request format with the model ID kimi-k3.

Is Kimi K3 free? Not for API use. Moonshot's pricing page lists $3.00 per million input tokens (cache miss), $0.30 per million (cache hit), and $15.00 per million output tokens, verified July 21, 2026. File upload and content extraction are listed as temporarily free. The kimi.com chat app's free-tier limits aren't detailed in Moonshot's public docs as of this writing.

What is the Kimi K3 API base URL? https://api.moonshot.ai/v1, per Moonshot's current quickstart documentation. Older guides may still show api.moonshot.cn/v1; that domain appears to have been the earlier endpoint, so confirm the live value before you build against it.

Is Kimi K3 open source? No. There's no K3 model listed on Moonshot's Hugging Face organization page as of this writing. Kimi K2.5 is the most recent model Moonshot has actually released with open weights.

Does Kimi K3 support images and video? Yes. Moonshot's own documentation describes K3 as natively multimodal with direct text, image, and video input support, rather than routing images through a separate vision model.

How does Kimi K3 compare to Claude Fable 5? On specs, availability, and price transparency, not benchmarks, since no official K3 benchmark suite exists yet. See our Kimi K3 vs Claude Fable 5 comparison for the full breakdown.

Where can I compare Kimi K3 to other models we track? Our model comparison hub tracks specs and pricing across the models we cover as they're verified, alongside the Kimi K3 comparison piece above.

Sources: Moonshot Kimi API quickstart, Kimi K3 pricing, Kimi API pricing overview, Moonshot API overview docs, Moonshot AI on Hugging Face